WO2024005137A1 - Control parameter generation method, program, recording medium, and control parameter-generating device - Google Patents
Control parameter generation method, program, recording medium, and control parameter-generating device Download PDFInfo
- Publication number
- WO2024005137A1 WO2024005137A1 PCT/JP2023/024162 JP2023024162W WO2024005137A1 WO 2024005137 A1 WO2024005137 A1 WO 2024005137A1 JP 2023024162 W JP2023024162 W JP 2023024162W WO 2024005137 A1 WO2024005137 A1 WO 2024005137A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- evaluation index
- index data
- control parameter
- control parameters
- driven object
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 151
- 238000011156 evaluation Methods 0.000 claims abstract description 344
- 238000010801 machine learning Methods 0.000 claims abstract description 60
- 238000005259 measurement Methods 0.000 claims abstract description 53
- 238000004519 manufacturing process Methods 0.000 claims description 116
- 238000012545 processing Methods 0.000 claims description 77
- 230000010365 information processing Effects 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 8
- 238000003754 machining Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 99
- 238000010586 diagram Methods 0.000 description 48
- 230000007717 exclusion Effects 0.000 description 18
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000006073 displacement reaction Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000007704 transition Effects 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 239000000047 product Substances 0.000 description 3
- 239000006185 dispersion Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005401 electroluminescence Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 230000032258 transport Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/4155—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by programme execution, i.e. part programme or machine function execution, e.g. selection of a programme
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D3/00—Control of position or direction
- G05D3/12—Control of position or direction using feedback
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P29/00—Arrangements for regulating or controlling electric motors, appropriate for both AC and DC motors
Definitions
- the present disclosure relates to a method of generating control parameters, and particularly to a method of generating control parameters used in production equipment.
- an object of the present disclosure is to provide a control parameter generation method etc. that can generate appropriate control parameters.
- a control parameter generation method includes generating a control parameter by a control parameter generation device, using the control parameter by a production device, using a sensor to check the state of the production device, and displaying at least a part of the control parameter by a display device. Generate control parameters.
- the production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. have things and.
- the sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position.
- the control parameter generation method includes the following method. That is, measurement data representing the position output from the sensor is acquired. Based on the measurement data, evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position is generated in association with a generation time of the evaluation index data. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters.
- the updated control parameters are output to the production equipment for storage in the memory.
- Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device.
- a program according to another aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
- a recording medium is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
- a control parameter generation device is a control parameter generation device that is used by a production device, a sensor confirms a state of the production device, and a display device generates a control parameter that displays at least a part of the control parameter.
- This is a parameter generation device.
- the production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- a driven object to be driven is a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- the sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position.
- the control parameter generation device includes an input section, a second control circuit, and an output section. The input unit acquires measurement data representing the position output from the sensor. Based on the measurement data, the second control circuit converts evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position into a generation time of the evaluation index data.
- the control parameters are updated using a machine learning model that is generated in association with each other and learns the relationship between the evaluation index data and the control parameters based on the evaluation index data.
- the output unit outputs the updated control parameters to the production device for storage in the memory.
- the output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device.
- the second control circuit When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
- a control parameter generation device generates the control parameter, a production device uses it, a sensor with a processing device checks the state of the production device, and a display device displays at least the state of the production device. Generate some control parameters to display.
- the production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores control parameters used by the control circuit to control the servo motor, and a driven object driven by the servo motor. and.
- the sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position.
- the position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output.
- the control parameter generation method includes the following method. Evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, is acquired in association with a generation time of the evaluation index data.
- control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters.
- the updated control parameters are output to the production equipment for storage in the memory.
- Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device.
- a program according to another aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
- a recording medium is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
- a control parameter generation device is used by a production device, a sensor with a processing device checks the state of the production device, and a display device generates control parameters that display at least a part of the control parameters.
- This is a control parameter generation device.
- the production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- a driven object to be driven is a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- the sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position.
- the position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output.
- the control parameter generation device includes an input section, a second control circuit, and an output section.
- the input unit may cause evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, to correspond to a generation time of the evaluation index data. Attach and obtain.
- the second control circuit updates the control parameter based on the evaluation index data using a machine learning model that learns a relationship between the evaluation index data and the control parameter.
- the output unit outputs the updated control parameters to the production device for storage in the memory.
- the output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device.
- the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
- control parameter generation method program, recording medium, and control parameter generation device according to the above aspects of the present disclosure, appropriate control parameters can be generated.
- FIG. 1 is a schematic diagram showing an overview of a control parameter generation system according to a first embodiment.
- FIG. 2 is a block diagram showing the configuration of the control parameter generation system according to the first embodiment.
- FIG. 3 is a schematic diagram showing an example of control parameters according to the first embodiment.
- FIG. 4 is a schematic diagram showing an example of the transition of the positional deviation of the driven object with respect to the target position.
- FIG. 5 is a data configuration diagram showing an example of measurement data according to the first embodiment.
- FIG. 6 is a schematic diagram showing how the machine learning model according to the first embodiment outputs control parameters based on the result of learning the relationship between evaluation index data and control parameters.
- FIG. 7 is a diagram illustrating an example of an image generated by the image generation unit according to the first embodiment.
- FIG. 8 is a diagram illustrating an example of an image generated by the image generation unit according to the first embodiment.
- FIG. 9 is a sequence diagram of the first control parameter adjustment process according to the first embodiment.
- FIG. 10 is a diagram showing a flowchart of the first control parameter adjustment process according to the first embodiment.
- FIG. 11 is a schematic diagram showing an overview of a control parameter generation system according to the second embodiment.
- FIG. 12 is a block diagram showing the configuration of a control parameter generation system according to the second embodiment.
- FIG. 13 is a sequence diagram of the second control parameter adjustment process according to the second embodiment.
- FIG. 14 is a diagram showing a flowchart of the second control parameter adjustment process according to the second embodiment.
- FIG. 15 is a schematic diagram showing an overview of the control parameter generation system according to the third embodiment.
- FIG. 16 is a block diagram showing the configuration of a control parameter generation system according to the third embodiment.
- FIG. 17 is a data configuration diagram showing an example of the data configuration of the adjustment history according to the third embodiment.
- FIG. 18 is a data configuration diagram showing an example of the data configuration of a threshold determination history according to the third embodiment.
- FIG. 19 is a diagram illustrating an example of an adjustment restart point designation image displayed by the display device according to the third embodiment.
- FIG. 20 is a diagram showing part 1 of a flowchart of the third control parameter adjustment process according to the third embodiment.
- FIG. 21 is a diagram showing a second flowchart of the third control parameter adjustment process according to the third embodiment.
- FIG. 22 is a schematic diagram showing an overview of the control parameter generation system according to the fourth embodiment.
- FIG. 17 is a data configuration diagram showing an example of the data configuration of the adjustment history according to the third embodiment.
- FIG. 18 is a data configuration diagram showing an example of the data configuration of a threshold determination history according to the third
- FIG. 23 is a block diagram showing the configuration of a control parameter generation system according to the fourth embodiment.
- FIG. 24 is a diagram showing part 1 of a flowchart of the fourth control parameter adjustment process according to the fourth embodiment.
- FIG. 25 is a diagram showing the second flowchart of the fourth control parameter adjustment process according to the fourth embodiment.
- FIG. 26 is a block diagram showing the configuration of a generation device according to a modification.
- FIG. 27 is a block diagram showing the configuration of a control parameter generation system according to a modification.
- FIG. 28 is a diagram showing vibration waveforms of the production equipment.
- control parameters used in production equipment equipped with a servo motor that drives a driven object have been manually adjusted.
- the number of control parameters used in such production equipment may reach 80, and the number of parameters that actually need to be adjusted may range from 30 to 60.
- the parameters to be adjusted differ depending on the drive target to be installed in the target production equipment, the application, and the like.
- Adjustment of such control parameters is necessary to suppress the generation of abnormal noise due to vibration, or to suppress the negative effect of vibration from the previous operation on the next operation when stopping an operation and starting the next operation. It is.
- Patent Document 1 proposes adjusting such control parameters using a machine learning model.
- skilled engineers adjust control parameters using all five human senses, such as checking abnormal noises by hearing and checking vibrations by sight and touch. Therefore, separate consideration is required as to what kind of data should be input to the machine learning model to effectively adjust the control parameters.
- control parameters can be effectively adjusted by inputting the evaluation index data into a machine learning model.
- the inventors investigated the cause of such a phenomenon and obtained the following findings.
- the production equipment etc. to be adjusted may not be constantly monitored by humans. For this reason, during the period when control parameters are being adjusted, people walk around the production equipment, etc. to be adjusted, and large trucks, etc., drive on roads around factories, etc. where the production equipment, etc. to be adjusted are installed. Or, if the vibration of the production equipment, etc. to be adjusted is having an adverse effect on the adjustment of control parameters, due to the opening and closing of the entrance door of the room where the production equipment, etc. to be adjusted is installed. It turns out that there is.
- Figure 28 is an example of the actual measurement data that served as the basis for the inventors to obtain the above knowledge.
- Vibration waveform of the production equipment showing actual measurement data of the displacement of the production equipment measured by a laser displacement meter when the production equipment is traveling and actual measurement data of the displacement of the production equipment measured by the laser displacement meter when the production equipment is not traveling. It is.
- the horizontal axis is the elapsed time
- the vertical axis is the amplitude of the displacement of the production equipment.
- the inventors discovered that even during a period when the amplitude of the production equipment is well within the threshold value when no large-sized trucks drive on the surrounding roads, the surrounding roads are It was noticed that the vibration of the production equipment was not within the threshold when the truck was running. The inventors discovered that when the above-mentioned phenomenon in which the vibration of the production equipment does not fall within the threshold range due to large trucks driving on the surrounding roads occurs during the period when the control parameters are being adjusted. It has been noticed that even after adjusting the control parameters, abnormal noises due to vibrations may still occur.
- the inventors acquired and analyzed further measured data, and discovered that the vibration of production equipment does not fall within the threshold range due to people walking around the production equipment, and that the production equipment is not installed. It was confirmed that the vibration of the production equipment did not fall within the threshold range due to the opening and closing of the door at the entrance to the room. Furthermore, even if such a phenomenon occurs during the period when the control parameters are being adjusted, abnormal noises due to vibration may still occur even after the control parameters have been adjusted. I found out that there is.
- a control parameter generation method includes generating a control parameter by a control parameter generation device, using the control parameter by a production device, using a sensor to check the state of the production device, and displaying at least a part of the control parameter by a display device. Generate control parameters.
- the production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. have things and.
- the sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position.
- the control parameter generation method includes the following method. That is, measurement data representing the position output from the sensor is acquired. Based on the measurement data, evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position is generated in association with a generation time of the evaluation index data. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters.
- the updated control parameters are output to the production equipment for storage in the memory.
- Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to a display device in the generation system.
- control parameter generation method by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate Control parameters can be generated.
- the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold.
- the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
- the machine learning model may learn the relationship between the evaluation index data and the control parameter using a Bayesian optimization algorithm.
- the machine learning model can use a Bayesian optimization algorithm to learn the relationship between evaluation index data and control parameters.
- the production device may include any one of a mounting device, a processing device, or a machining device.
- control parameters can be generated for a production device that is any one of a mounting device, a processing device, or a machining device.
- a program according to one aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
- the newly generated evaluation index data is generated during at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold.
- the period for re-adjusting the control parameters can be shortened compared to the case where the control parameters are re-adjusted from the beginning.
- a recording medium is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
- control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate control parameters can be determined without the need of skilled engineers and in a relatively short time. can be generated.
- the newly generated evaluation index is generated in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold.
- the period of re-adjustment can be shortened compared to the case of re-adjusting the control parameters from the beginning.
- a control parameter generation device is a control parameter generation device that is used by a production device, a sensor confirms a state of the production device, and a display device generates a control parameter, at least a part of which is displayed. It is a generating device.
- the production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- a driven object to be driven The control parameter generation device further includes an input section, a second control circuit, and an output section.
- the sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position.
- the input unit acquires measurement data representing the position output from the sensor.
- the second control circuit converts evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position into a generation time of the evaluation index data.
- the control parameters are updated using a machine learning model that is generated in association with each other and learns the relationship between the evaluation index data and the control parameters based on the evaluation index data.
- the output unit outputs the updated control parameters to the production device for storage in the memory.
- the output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device.
- the second control circuit When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
- control parameters can be updated based on the evaluation index data indicating the vibration after the time when the driven object reaches the permissible position. , it is possible to generate appropriate control parameters.
- the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold.
- a control parameter generation device generates the control parameter, a production device uses it, a sensor with a processing device checks the state of the production device, and a display device displays at least the state of the production device. Generate some control parameters to display.
- the production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. and a sensor with a processing device.
- the sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position.
- the position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output.
- the control parameter generation method includes the following method. That is, evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position, which is output from the sensor with the processing device, is obtained in association with the generation time of the evaluation index data. do.
- control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters.
- the updated control parameters are output to the production equipment for storage in the memory.
- Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device in the generation system.
- control parameter generation method by updating the control parameters based on the evaluation index data indicating the vibration after the time when the driven object reaches the permissible position, the , it is possible to generate appropriate control parameters.
- the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold.
- the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
- the machine learning model may learn the relationship between the evaluation index data and the control parameter using a Bayesian optimization algorithm.
- the machine learning model can use a Bayesian optimization algorithm to learn the relationship between evaluation index data and control parameters.
- the evaluation index data may be generated in the sensor with a processing device based on the position of the driven object measured by the sensor with a processing device.
- the evaluation index data can be generated based on the position of the driven object measured by the sensor with the processing device.
- the senor with a processing device may include an image processing device.
- the evaluation index data can be generated based on pixel processing.
- the production device may include any one of a mounting device, a processing device, or a machining device.
- control parameters can be generated for a production device that is any one of a mounting device, a processing device, or a machining device.
- a program according to one aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
- a recording medium is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
- control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate control parameters can be determined without the need of skilled engineers and in a relatively short time. can be generated.
- the newly generated evaluation index is generated in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold.
- the period of re-adjustment can be shortened compared to the case of re-adjusting the control parameters from the beginning.
- a control parameter generation device is used by a production device, a sensor with a processing device checks the state of the production device, and a display device generates control parameters that display at least a part of the control parameters.
- This is a control parameter generation device.
- the production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- a driven object to be driven is a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor.
- the sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position.
- the position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output.
- the control parameter generation device includes an input section, a second control circuit, and an output section.
- the input unit may cause evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, to correspond to a generation time of the evaluation index data. Attach and obtain.
- the second control circuit updates the control parameter based on the evaluation index data using a machine learning model that learns a relationship between the evaluation index data and the control parameter.
- the output unit outputs the updated control parameters to the production device for storage in the memory.
- the output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device.
- the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
- control parameter generation device by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate Control parameters can be generated.
- the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold.
- This control parameter generation system 1 is a system that generates control parameters used in a production apparatus 20 that includes a servo motor 23 that drives a driven object 24 .
- FIG. 1 is a schematic diagram showing an overview of a control parameter generation system 1 according to the first embodiment.
- FIG. 2 is a block diagram showing the configuration of the control parameter generation system 1.
- the control parameter generation system 1 includes a generation device 10, a production device 20, and a sensor 30.
- the sensor 30 may be provided separately from the production device 20, or may be a component included in the production device 20. Also. , the sensor 30 may be an inspection device independent of the production device 20 or a part thereof.
- the production device 20 is a device used to produce equipment, and processes, mounts, transports, etc. the equipment.
- the production device 20 is installed, for example, on a production line of a factory.
- the production equipment 20 is, for example, an LED (light emitting diode) bonder, a mounting machine, a processing machine, a take-out robot, or the like.
- the production device 20 includes a memory 21, a control circuit 22, a servo motor 23, and a driven object 24.
- the operation of the servo motor 23 is controlled by the control circuit 22 and drives the object 24 to be driven.
- the servo motor 23 may be, for example, a rotary motor or a linear motor.
- the driven object 24 is an object driven by the servo motor 23.
- the driven object 24 is, for example, a processing object to be processed in the production apparatus 20, a mounting object to be mounted, a conveyance object to be transported, and the like.
- the driven object 24 may be, for example, an arm that holds objects handled in the production apparatus 20, or a tray that transports these objects.
- the control circuit 22 controls the servo motor 23 by outputting a command to the servo motor 23 to position the driven object 24 at a predetermined target position.
- the command that the control circuit 22 outputs to the servo motor 23 may be, for example, a position command that commands the position of the servo motor 23 or the driven object 24, or may be, for example, a torque command that commands the torque of the servo motor 23. There may be.
- the control circuit 22 controls the servo motor 23 based on the control parameters stored in the memory 21. In other words, the control circuit 22 uses the control parameters stored in the memory 21 when controlling the servo motor 23.
- the memory 21 stores control parameters used by the control circuit 22 to control the servo motor 23.
- the control parameters stored in the memory 21 are the control parameters output from the generation device 10.
- FIG. 3 is a schematic diagram showing an example of control parameters stored in the memory 21.
- the control parameters stored in the memory 21 include, for example, parameters a1 and a2 for adjusting the vibration frequency of the driven object 24, parameters b1 and b2 for adjusting the speed increase of the driven object 24, and parameters b1 and b2 for adjusting the speed of the driven object 24.
- the parameters include parameters c1 and c2 for adjusting the depth of a singular point in the vibration characteristics of the object 24, parameters d1 and d2 for adjusting the vibration amplitude of the driven object 24, and the like.
- control parameters include, for example, parameters b1 and b2 that adjust speed, parameters c1 and c2 that adjust the depth of singularity in vibration characteristics, and parameters d1 and d2 that adjust vibration amplitude.
- the sensor 30 measures the position of the driven object 24. Then, the sensor 30 outputs to the generation device 10 measurement data representing the position of the driven object 24 after the time when the driven object 24 reaches an allowable position where it can be evaluated that the driven object 24 has reached a predetermined target position.
- FIG. 4 is a schematic diagram showing an example of the transition of the positional deviation of the driven object 24 with respect to the target position when the production apparatus 20 drives the driven object 24 to the target position.
- the horizontal axis indicates time
- the vertical axis indicates the positional deviation of the driven object 24 with respect to the target position.
- an allowable position refers to a position where the positional deviation from the target position is within the required accuracy.
- the time when the driven object 24 reaches the permissible position (hereinafter also referred to as "settling time") at which it can be evaluated that the target position has been reached is the time when the driven object 24 reaches the permissible position. This refers to the time when the permissible position is finally reached (time sett_t in FIG. 4) when the permissible position does not deviate again after reaching the permissible position.
- FIG. 5 is a data configuration diagram showing an example of measurement data output by the sensor 30.
- the measurement data is, for example, data in which the elapsed time [ms] after the time when the permissible position is reached and the deviation amount [mm] from the target position are associated in one-to-one correspondence.
- the generation device 10 generates control parameters (hereinafter also referred to as “updating control parameters") for updating the control parameters stored in the memory 21 (hereinafter also referred to as “stored control parameters"). More specifically, the generation device 10 sequentially acquires measurement data output from the sensor 30 as a result of the production device 20 driving the driven object 24 to the target position using the stored control parameters, and uses the acquired measurement data. Update control parameters are sequentially generated based on the data, and the generated update control parameters are sequentially output to the production apparatus 20.
- the generation device 10 includes an input section 11, an output section 12, a control circuit 13, a machine learning model 14, an operation reception section 15, an image generation section 16, and a display section 17. Be prepared.
- the generation device 10 is realized, for example, in a computer device including a processor, a memory, and an input/output interface, when the processor executes a program stored in the memory.
- a computer device including a processor, a memory, and an input/output interface, when the processor executes a program stored in the memory.
- a computer device is, for example, a personal computer.
- the input unit 11 acquires the measurement data output from the sensor 30.
- the input unit 11 may be, for example, an electronic component or an electronic circuit that inputs a signal.
- the control circuit 13 generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data acquired by the input unit 11, and calculates the evaluation index data based on the generated evaluation index data.
- the control parameters are updated using the machine learning model 14 that learns the relationship with the control parameters.
- the evaluation index data generated by the control circuit 13 is any one of the following (1) to (3) or (4). That is, (1) The sum total of the area surrounded by the deviation waveform indicating the vibration of the driven object 24 after the settling time and the reference axis (hereinafter also referred to as “total area evaluation index data”). (2) Degree of variation in the deviation waveform indicating the vibration of the driven object 24 after the settling time (hereinafter also referred to as “dispersion degree evaluation index data”). (3) Effective value of the deviation waveform indicating the vibration of the driven object 24 after the settling time (hereinafter also referred to as "effective value evaluation index data”). any one of the following, or (4) Time from the time when the driven object 24 starts to be driven to the settling time (hereinafter also referred to as "settling time evaluation index data”). It is.
- the evaluation index data generated by the control circuit 13 is specified by the user using the control parameter generation system 1.
- the starting point sett_t of the integral interval is the settling time
- the end point cal_e of the integral interval is the time set by the user. Note that although the explanation will be given here assuming that the starting point sett_t of the integral interval is the settling time, the starting point sett_t of the integral interval may be the settling time or any time after the settling time.
- cal_e is, for example, the final position when the positional deviation of the driven object 24 with respect to the target position reaches within 1/10 of the required accuracy and then the positional deviation no longer deviates from within 1/10 of the required accuracy. It may be set at the time when the deviation reaches within 1/10 of the required accuracy, or for example, it may be set at the time when the number of data included in the measurement data acquired by the input unit 11 reaches a predetermined number. , it may be set to a time when a specified time has elapsed from the settling time.
- the effective value evaluation index data generated by the control circuit 13 is calculated by the formula shown in (Formula 5).
- the machine learning model 14 is a machine learning model that learns the relationship between evaluation index data and control parameters, and learns the relationship between evaluation index data and control parameters each time evaluation index data is sequentially input from the control circuit 13. It learns and sequentially outputs control parameters to the control circuit 13 based on the learning results.
- the machine learning model 14 uses a Bayesian optimization algorithm to calculate evaluation index data and control parameters corresponding to the evaluation index data each time the evaluation index data is sequentially input from the control circuit 13. (hereinafter also referred to as "corresponding control parameters"), and outputs to the control circuit 13 the control parameters predicted to have the smallest evaluation index data based on the learning results.
- FIG. 6 is a schematic diagram showing how the machine learning model 14 learns the relationship between evaluation index data and corresponding control parameters, and outputs the control parameters that are predicted to cause the smallest evaluation index data based on the learned results. It is a diagram.
- the horizontal axis is the value of the control parameter
- the vertical axis is the value of the evaluation index data.
- the points plotted with circles indicate the values of the corresponding control parameters of the evaluation index data for each value of the evaluation index data input into the machine learning model 14 in the past.
- the machine learning model 14 uses a Bayesian optimization algorithm to predict the relationship between evaluation index data and corresponding control parameters within a certain range, and uses the predicted relationship to A control parameter that is predicted to have the smallest evaluation index data is output.
- the output unit 12 outputs the control parameters updated by the control circuit 13 to the production device 20 in order to store them in the memory 21.
- the output unit 12 may be, for example, an electronic component or an electronic circuit that outputs a signal.
- the operation accepting unit 15 accepts an input operation to the generating device 10 by a user using the control parameter generating system 1.
- the operation receiving unit 15 may be, for example, an input device such as a button or a keyboard.
- the display unit 17 displays images provided to the user who uses the control parameter generation system 1.
- the display unit 17 may be, for example, a liquid crystal monitor, an organic EL (Electro Luminescence) monitor, or a CRT (Cathode Ray Tube) monitor.
- the image generation unit 16 generates an image to be displayed on the display unit 17.
- FIG. 7 is an example of an image generated by the image generation unit 16.
- FIG. 7 shows how the control parameter generation system 1 sends evaluation index data generated by the control circuit 13 to a user who uses the control parameter generation system 1 when starting a first control parameter adjustment process to be described later.
- This is an example of an image (hereinafter also referred to as "evaluation index data designation image") that prompts the user to specify whether or not to use the evaluation index data.
- the image generation unit 16 may be, for example, a graphic board.
- the generation device 10 displays an evaluation index data designation image on the display unit 17, thereby prompting the user to designate settling time evaluation index data as the evaluation index data generated by the control circuit 13. , or any one of total area evaluation index data, variation degree evaluation index data, and effective value evaluation index data (hereinafter, these three evaluation index data are also referred to as "vibration evaluation index data"). Encourage one person to specify one.
- the user specifies the settling time evaluation index data as the evaluation index data, for example, by placing a check next to "settling time” in the evaluation index data designation image.
- the user can select area sum evaluation index data as evaluation index data by checking the box next to "vibration” and putting a circle next to "area” in the evaluation index data specification image.
- the user can select the degree of variation evaluation index data as evaluation index data by checking the box next to "vibration” and putting a circle next to "dispersion” in the evaluation index data specification image.
- the user can select the effective value as evaluation index data by checking the box next to "vibration” and putting a circle next to "RMS (effective value)" in the evaluation index data specification image.
- RMS effective value
- the user can, for example, check the boxes next to "Settling time” and "Vibration" in the evaluation index data specification image to select the settling time evaluation index data and vibration as evaluation index data.
- the control circuit 13 may set the value of the settling time evaluation index data to be the sum of the value of the settling time evaluation index data and the value of the vibration evaluation index data.
- vibration evaluation index data, and the value of the evaluation index data may be the sum of the values obtained by multiplying each value by a predetermined weighting coefficient.
- FIG. 8 is an example of an image generated by the image generation unit 16.
- FIG. 8 shows that while the control parameter generation system 1 is executing the first control parameter adjustment process, which will be described later, a user using the control parameter generation system 1 is asked to perform the first control parameter adjustment process.
- This is an example of an image that presents the status of the value of evaluation index data (hereinafter also referred to as a "situation presentation image").
- the horizontal axis is the number of control parameters sequentially output by the generating device 10
- the vertical axis is the value of evaluation index data corresponding to each of the control parameters sequentially output by the generating device 10.
- the solid line indicates the transition of the minimum value of the evaluation index data corresponding to the control parameter output by the generation device 10.
- the interruption icon indicates, for example, that the value of the evaluation index data corresponding to the control parameter output by the generation device 10 has become sufficiently small as a result of the user visually confirming the situation presentation image, so that the first control parameter is no longer used. This icon allows you to interrupt the first control parameter adjustment process by clicking the interrupt icon when it is determined that there is no need to continue the adjustment process.
- the restart icon indicates, for example, that the first control parameter adjustment process is performed because the value of the evaluation index data corresponding to the control parameter output by the generation device 10 has not become sufficiently small as a result of the user viewing the situation presentation image. This icon allows you to continue the first control parameter adjustment process by clicking the restart icon when it is determined that it is necessary to continue the process.
- the control parameter generation system 1 executes a first control parameter adjustment process that adjusts the control parameters stored in the memory 21 to appropriate values.
- the first control parameter adjustment process is started, for example, when a user using the control parameter generation system 1 performs an operation on the generation device 10 to start the first control parameter adjustment process.
- FIG. 9 is a sequence diagram of the first control parameter adjustment process
- FIG. 10 is a flowchart of the first control parameter adjustment process.
- the generation device 10 starts a predetermined program for executing the first control parameter adjustment process (step S10).
- the output unit 12 When the predetermined program is started, the output unit 12 outputs the initial values of the control parameters to the memory 21 (step S15). At this time, the output unit 12 may, for example, output an initial value of the control parameter consisting of a predetermined value, or may output an initial value of the control parameter consisting of a value specified by the user. Alternatively, for example, the initial value of the control parameter may be output, which is a value calculated by a calculation method specified by the user.
- the memory 21 stores the control parameters output from the output unit 12 (step S20). Then, the control circuit 22 generates a command to position the driven object 24 to a predetermined target position based on the control parameters stored in the memory 21 and outputs it to the servo motor 23. Thereby, the control circuit 22 controls the servo motor 23 (step S25). Then, the servo motor 23 drives the driven object 24 based on the command output from the control circuit 22 (step S30).
- the sensor 30 measures the position of the driven object 24 (step S35), and moves the driven object 24 to an allowable position where it can be evaluated that it has reached a predetermined target position. Measurement data representing the position of the driven object 24 after the time reached is output to the input unit 11 . Then, the input unit 11 acquires the measurement data output from the sensor 30 (step S40).
- the control circuit 13 When the input unit 11 acquires the measurement data, the control circuit 13 generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data (step S45). Then, the control circuit 13 checks whether the generation of evaluation index data satisfies a predetermined condition (step S50).
- the predetermined conditions include, for example, the condition that the number of times the evaluation index data has been updated (that is, the number of times it has been generated) has reached a predetermined number, the condition that a predetermined time has elapsed since the start of adjustment of the control parameters, and the like.
- the predetermined number of times and the predetermined time may be determined in advance, or may be specified by the user using the control parameter generation system 1.
- step S50 if the generation of evaluation index data does not satisfy the predetermined condition (step S50: No), the control circuit 13 outputs the generated evaluation index data to the machine learning model 14. Then, the machine learning model 14 learns the relationship between the evaluation index data and the corresponding control parameter (step S55), and outputs the control parameter that is predicted to have the smallest value of the evaluation index data. Then, the control circuit 13 updates the control parameters previously output by the output unit 12 with the control parameters newly output from the machine learning model 14 (step S60). Then, the output unit 12 transmits the control parameters newly updated by the control circuit 13 to the memory 21 (step S65).
- step S65 the control parameter generation system 1 proceeds to the process of step S20, and repeats the process from step S20 onwards.
- step S50 if the generated evaluation index data satisfies the predetermined condition (step S50: Yes), the control parameter generation system 1 ends the first control parameter adjustment process.
- control parameter generation system 1 configured as described above repeatedly generates control parameters by executing the first control parameter adjustment process until the generation of evaluation index value data satisfies a predetermined condition. Adjust the control parameters.
- control parameter generation system 1 appropriate control parameters can be generated in a relatively short time without the need for skilled engineers.
- the control parameter generation system 1 was an example of a system in which the sensor 30 measures the position of the driven object 24 and outputs measurement data, and the generation device 10 acquires the measurement data output from the sensor 30.
- the sensor with the processing device according to the second embodiment measures the position of the driven object 24 and outputs evaluation index data.
- the generation device according to Embodiment 2 is an example of a system that acquires evaluation index data output from the sensor with a processing device according to the second embodiment.
- control parameter generation system according to Embodiment 2, the same components as those in control parameter generation system 1 are given the same reference numerals as those in control parameter generation system 1, and the detailed explanation thereof will be omitted. The differences from System 1 will be mainly explained.
- FIG. 11 is a schematic diagram showing an overview of a control parameter generation system 1A according to the second embodiment.
- FIG. 12 is a block diagram showing the configuration of the control parameter generation system 1A.
- the control parameter generation system 1A differs from the control parameter generation system 1 according to the first embodiment in that the generation device 10 is changed to a generation device 10A, and the sensor 30 is changed to a sensor with a processing device 30A. configured. Further, as shown in FIG. 12, the generation device 10A is configured by changing the input unit 11 from the generation device 10 to an input unit 11A, and changing the control circuit 13 to a control circuit 13A.
- the sensor 30A with a processing device measures the position of the driven object 24. Then, the processing device-equipped sensor 30A generates measurement data representing the position of the driven object 24 after the time when the driven object 24 reaches a permissible position where it can be evaluated that the driven object 24 has reached a predetermined target position. Furthermore, the sensor with processing device 30A generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the generated measurement data, and outputs the generated evaluation index data to the generation device 10A.
- the sensor with processing device 30A is, for example, a computer device including a sensing device that measures the position of the driven object 24, a processor, a memory, and an input/output interface, in which the processor executes a program stored in the memory. It is realized by Such a computer device is, for example, a personal computer (personal computer).
- the processing device-equipped sensor 30A may include, for example, an image processing device.
- the sensor with processing device 30A may measure the position of the driven object 24 by performing image processing on the image of the driven object 24 using an image processing device, for example.
- the input unit 11A acquires the evaluation index data output from the sensor with processing device 30A.
- the control circuit 13A updates the control parameters using the machine learning model 14 that learns the relationship between the evaluation index data and the control parameters based on the evaluation index data acquired by the input unit 11A.
- the control parameter generation system 1A performs a second control parameter adjustment process in which a part of the process is changed from the first control parameter adjustment process according to the first embodiment.
- FIG. 13 is a sequence diagram of the second control parameter adjustment process
- FIG. 14 is a flowchart of the second control parameter adjustment process.
- the second control parameter adjustment process includes the process from step S110 to step S165.
- the process from step S110 to step S130 and the process from step S150 to step S165 are the process from step S10 to step S30 in the first control parameter adjustment process, respectively.
- the process is similar to the process from step S50 to step S65.
- step S135 the processing from step S135 to step S145 will be mainly explained.
- the sensor with processing device 30A measures the position of the driven object 24 (step S135), and the time when the driven object 24 reaches an allowable position at which it can be evaluated that it has reached a predetermined target position. Measurement data representing the subsequent position of the driven object 24 is generated. Then, the sensor with processing device 30A generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the generated measurement data (step S140), and inputs the generated evaluation index data to the input section 11A. Output to. Then, the input unit 11A obtains the evaluation index data output from the processing device-equipped sensor 30A (step S145).
- step S145 Upon completion of the process in step S145, the control parameter generation system 1A proceeds to the process in step S150.
- step S150 if the generation of evaluation index data satisfies the predetermined conditions (step S150: Yes), the control parameter generation system 1A ends the second control parameter adjustment process.
- control parameter generation system 1A having the above configuration can generate evaluation index value data by executing the second control parameter adjustment process.
- Control parameters are adjusted by repeatedly generating control parameters until predetermined conditions are met.
- control parameter generation system 1A can generate appropriate control parameters in a relatively short period of time without requiring a skilled engineer.
- the control parameter generation system is an example of a system that can readjust the control parameters described below. That is, the control parameter generation system adjusts the control parameters by repeatedly generating the control parameters until the generation of the evaluation index data satisfies a predetermined condition. If the adjusted evaluation index data is not satisfactory to the user using the control parameter generation system, the control parameter generation system is based on evaluation index data generated during a time period excluding the time period specified by the user. Then readjust the control parameters.
- control parameter generation system 1B According to Embodiment 3, the same components as those in the control parameter generation system 1 will be given the same reference numerals as they have already been explained, and detailed explanation thereof will be omitted. The differences from generation system 1 will be mainly explained.
- FIG. 15 is a schematic diagram showing an overview of the control parameter generation system 1B according to the third embodiment.
- FIG. 16 is a block diagram showing the configuration of the control parameter generation system 1B.
- the control parameter generation system 1B is configured by changing the generation device 10 from the control parameter generation system 1 according to Embodiment 1 to a generation device 10B, and adding a display device 40.
- the display device 40 may be provided separately from the generation device 10B, or may be included in the generation device 10B. Further, the display device 40 and the display section 17 may be the same. In a case where the display device 40 is included as part of the display unit 17, for example, the display unit 17 may include a plurality of display devices, and the display device 40 may be one of them. Further, information shown by the display device 40 may be displayed on the display unit 17.
- the generation device 10B is configured by changing the output unit 12 from the generation device 10 to an output unit 12B and changing the control circuit 13 to a control circuit 13B.
- the control circuit 13B performs operations similar to those performed by the control circuit 13, and further performs the following operations.
- the control circuit 13B adjusts the control parameters by repeatedly generating control parameters using the control parameter generation system 1B until the generation of evaluation index data satisfies a predetermined condition. Thereafter, the user using the control parameter generation system 1B specifies at least one time period and instructs to restart updating of the control parameters.
- the machine learning model 14 is executed based on the evaluation index data that is newly generated in at least one time period and the evaluation index data that has already been generated, except for the evaluation index data that is newly generated. to re-update the updated control parameters.
- control circuit 13B When generating evaluation index data, the control circuit 13B generates the evaluation index data in association with the generation time of the evaluation index data. Then, the control circuit 13B updates the adjustment history based on the generated evaluation index data and the generation time of the evaluation index data.
- FIG. 17 is a data configuration diagram showing an example of the data configuration of the adjustment history updated by the control circuit 13B.
- the adjustment history is configured by associating, for example, the number of updates, a time stamp, the best evaluation value, and the control parameter value at the time of the best evaluation value.
- the number of updates is the number of times the control circuit 13B has generated the control parameters, that is, the number of times the control circuit 13B has updated the control parameters after the control parameter generation system 1B starts adjusting the control parameters.
- the time stamp is the generation time of the evaluation index data when the control circuit 13B generated the evaluation index data the number of times indicated by the corresponding number of updates.
- the best evaluation value is the value of the best evaluation index data (here, the minimum (the value of the evaluation index data).
- the control parameter value at the time of the best evaluation value is the value of the control parameter when the control circuit 13B calculates the corresponding best evaluation value.
- control circuit 13B After generating the evaluation index data, the control circuit 13B determines whether the vibration of the driven object 24 indicated by the generated evaluation index data exceeds a predetermined threshold. Then, the control circuit 13B updates the threshold determination history based on the results of the investigation.
- FIG. 18 is a data configuration diagram showing an example of the data configuration of the threshold determination history updated by the control circuit 13B.
- the threshold determination history is configured by associating, for example, the number of updates, a timestamp, a ratio to the threshold, and a determination result.
- the number of updates and the time stamp are the same as the number of updates and the time stamp in the adjustment history shown in FIG.
- the ratio to the threshold is the ratio of the vibration of the driven object 24 indicated by the evaluation index data to the predetermined threshold when the evaluation index data is generated the number of times indicated by the corresponding number of updates.
- the determination result is information that indicates "True” when the ratio to the corresponding threshold value exceeds 1, and indicates “False” when the ratio does not exceed 1.
- the determination result is, for example, a 1-bit signal that has a logical value of "1" when the ratio to the corresponding threshold value exceeds 1, and has a logical value of "0" when it does not exceed 1.
- the control circuit 13B refers to the threshold value determination history and associates the determination result with "Fault".
- the evaluation index data that was updated for the first update number is specified as good evaluation index data, and the evaluation index data that was updated for the first update number for which the determination result is "True" is specified as poor evaluation index data.
- the excellent evaluation index data is evaluation index data in which the vibration of the driven object 24 indicated by the evaluation index data does not exceed a predetermined threshold value
- the poor evaluation index data is the This is evaluation index data indicating that the vibration of the object 24 exceeds a predetermined threshold.
- control circuit 13B calculates the expected time required to update the evaluation index data again as many times as the number of excellent evaluation index data.
- the output unit 12B performs the same operations as the output unit 12, and further performs the following operations.
- the output unit 12B outputs time data indicating one or more time periods in which the vibration of the driven object 24 indicated by the evaluation index data exceeds a predetermined threshold value to the display device 40.
- the output unit 12B outputs time data to the display device 40 by outputting the adjustment history and the threshold value determination history to the display device 40.
- the output unit 12B outputs additional information indicating the expected time calculated by the control circuit 13B to the display device 40.
- the display device 40 generates and displays an image based on the data and information output from the output unit 12B.
- the display device 40 then receives input from a user using the control parameter generation system 1B based on the displayed image.
- FIG. 19 is an example of an image displayed by the display device 40.
- FIG. 19 shows an image (hereinafter referred to as This is an example of an image (also referred to as an "exclusion target section designation image").
- the horizontal axis is the number of updates of the control parameters or the elapsed time (adjustment time) since the start of adjustment of the control parameters, and the vertical axis is the best evaluation value.
- the exclusion target section is an evaluation index that is recommended to be excluded from evaluation index data to be retrained when relearning a machine learning model for a user using the control parameter generation system 1B.
- the exclusion target section is a time period in which one or more pieces of defective evaluation index data are continuously generated.
- the expected adjustment time is the expected readjustment time when the control parameters are readjusted.
- the expected adjustment time is the expected time calculated by the control circuit 13.
- the broken line in FIG. 19 is a virtual line indicating the transition of the best evaluation value when the control parameter adjustment progresses ideally, and is not actually displayed in the adjustment restart point designation image.
- the user can apply machine learning to the evaluation index data corresponding to all sections of the excluded sections to be displayed.
- the user can apply machine learning to the evaluation index data corresponding to all sections of the excluded sections to be displayed.
- it can be excluded from evaluation index data targeted for relearning.
- the user can, for example, click the button next to "Select Exclusion Target Section" on the exclusion target section designation image, and then select (for example, click) the desired exclusion target section to select the desired exclusion target section.
- the evaluation index data corresponding to the exclusion target section can be excluded from the evaluation index data to be retrained when retraining the machine learning model.
- the control circuit 13B re-identifies the poor evaluation index data corresponding to the exclusion target section that was not selected as the desired exclusion target section as good evaluation index data.
- the control circuit 13B may recalculate the expected time using the newly specified excellent evaluation index data.
- the display device 40 may then update the displayed expected adjustment time with the recalculated expected time.
- the control parameter generation system 1B performs a third control parameter adjustment process in which a part of the process is changed from the first control parameter adjustment process according to the first embodiment.
- the third control parameter adjustment process includes the process from step S210 to step S297.
- the process from step S210 to step S240 and the process from step S250 to step S265 are the process from step S10 to step S40 in the first control parameter adjustment process, respectively.
- the process is similar to the process from step S50 to step S65.
- step S245 to step S247 the processing from step S270 to step S297 will be mainly described.
- step S240 the control circuit 13B generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data in association with the generation time of the evaluation index data (step S245). Then, the control circuit 13B updates the adjustment history based on the generated evaluation index data and the generation time of the evaluation index data (step S246).
- control circuit 13B After generating the evaluation index data, the control circuit 13B determines whether the vibration of the driven object 24 indicated by the generated evaluation index data exceeds a predetermined threshold. Then, the control circuit 13B updates the threshold determination history based on the results of the investigation (step S247).
- control parameter generation system 1B proceeds to the process in step S250.
- step S250 if the generation of evaluation index data satisfies a predetermined condition (step S250: Yes), the control circuit 13B calculates the expected time (step S270). Then, the output unit 12B outputs time data indicating one or more time periods in which the vibration of the driven object 24 indicated by the evaluation index data exceeds a predetermined threshold value to the display device 40 together with additional information indicating the expected time (Ste S271). Then, the display device 40 generates and displays an exclusion target section designation image based on these data and information (step S272).
- the display device 40 After displaying the exclusion target section designation image, the display device 40 accepts input by the user using the control parameter generation system 1B based on the displayed exclusion target section designation image (step S275).
- step S280 If the input received by the display device 40 is an input specifying an exclusion target section when readjusting the control parameters (step S280: Yes), the control circuit 13B controls the learning of the machine learning model 14. After initializing the degree, that is, rewinding to the start point of the third control parameter adjustment process, the machine learning model 14 is provided with excellent evaluation index data and the relationship between the evaluation index data and the corresponding control parameter. is re-learned (step S285).
- unselected excellent evaluation index data refers to data that has not yet been selected in a loop process that proceeds from the process of step S286: Yes to the process of step S297, which will be described later, and then returns to the process of step S286. This is excellent evaluation index data.
- step S286 if there is unselected excellent evaluation index data (step S286: Yes), the control circuit 13B selects one of the unselected excellent evaluation index data (Ste S287). Then, the control circuit 13B updates the control parameters with the control parameters corresponding to the selected excellent evaluation index data (step S288). Then, the output unit 12B transmits the control parameters newly updated by the control circuit 13B to the memory 21 (step S289).
- control parameter generation system 1B proceeds to the process in step S290.
- step S290 to step S296 is similar to the processing from step S220 to step S246, respectively.
- step S296 the machine learning model 14 learns the relationship between the generated evaluation index data and the corresponding control parameter (step S297).
- control parameter generation system 1B proceeds to the process in step S286.
- step S280 if the input received by the display device 40 is not an input to specify an exclusion target section when readjusting the control parameters (step S280: No), in the process of step S286 If there is no unselected excellent evaluation index data (step S286: No), the control parameter generation system 1B ends the third control parameter adjustment process.
- the control parameter generation system 1B having the above configuration is configured such that even if it is necessary to re-adjust the control parameters, the control parameter generation system 1B generates a new one in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold.
- the control parameters are re-updated based on the already generated evaluation index data other than the evaluation index data that has been generated. By doing so, it is possible to shorten the period for re-adjusting the control parameters compared to the case where the control parameters are re-adjusted from the beginning.
- the control parameter generation system 1B was an example of a system in which the sensor 30 measures the position of the driven object 24 and outputs measurement data, and the generation device 10B acquires the measurement data output from the sensor 30.
- the sensor with the processing device according to the fourth embodiment measures the position of the driven object 24 and outputs evaluation index data.
- the generation device according to Embodiment 4 is an example of a system that acquires evaluation index data output from the sensor with a processing device according to the fourth embodiment.
- control parameter generation system according to Embodiment 4, the same components as those in the control parameter generation system 1B will be given the same reference numerals as they have already been explained, and detailed explanation thereof will be omitted. The explanation will focus on the differences from system 1B.
- FIG. 22 is a schematic diagram showing an overview of the control parameter generation system 1C according to the fourth embodiment.
- FIG. 23 is a block diagram showing the configuration of the control parameter generation system 1C.
- the control parameter generation system 1C differs from the control parameter generation system 1B according to the third embodiment in that the generation device 10B is changed to a generation device 10C, and the sensor 30 is changed to a sensor with a processing device 30C. configured. Further, as shown in FIG. 23, the generation device 10C is configured by changing the input section 11 from the generation device 10B to an input section 11C, and changing the control circuit 13B to a control circuit 13C.
- the sensor with processing device 30C performs the same operations as the sensor with processing device 30A, and further performs the following operations.
- the processing device equipped sensor 30C When the sensor with processing device 30C generates evaluation index data, it generates the evaluation index data in association with the generation time of the evaluation index data. Then, when outputting the generated evaluation index data to the generation device 10C, the processing device equipped sensor 30C outputs the generation time of the generated evaluation index data in association with the evaluation index data.
- the input unit 11C performs the same operations as the input unit 11, and further performs the following operations.
- the input unit 11C acquires the evaluation index data output from the sensor with processing device 30C in association with the generation time of the evaluation index data.
- the control circuit 13C performs the same operations as the control circuit 13B, and further performs the following operations.
- the control circuit 13C When the input unit 11C does not acquire the measurement data from the sensor with processing device 30C and acquires the evaluation index data in association with the generation time of the evaluation index data, the control circuit 13C generates the evaluation index data based on the measurement data. Instead of generating the evaluation index data in association with the generation time of the evaluation index data, the evaluation index data acquired by the input unit 11C is acquired in association with the generation time of the evaluation index data.
- the control parameter generation system 1C performs a fourth control parameter adjustment process in which a part of the process is changed from the third control parameter adjustment process according to the third embodiment.
- the fourth control parameter adjustment process includes the process from step S310 to step S397.
- the process from step S310 to step S330 and the process from step S347 to step S397 are the process from step S210 to step S230 in the third control parameter adjustment process, respectively.
- the processing is similar to the processing from step S247 to step S297.
- step S335 the processing from step S335 to step S346 will be mainly described.
- the sensor with processing device 30C measures the position of the driven object 24 (step S335), and determines the time when the driven object 24 reaches an allowable position at which it can be evaluated that it has reached a predetermined target position. Measurement data representing the subsequent position of the driven object 24 is generated. Then, based on the generated measurement data, the processing device-equipped sensor 30C generates evaluation index data indicating the vibration of the driven object 24 after the settling time in association with the generation time of the evaluation index data (step S340 ), and outputs the generated evaluation index data to the input unit 11C in association with the generation time of the evaluation index data.
- the input unit 11C obtains the evaluation index data output from the processing device equipped sensor 30C in association with the generation time of the evaluation index data (step S345). Then, the control circuit 13C updates the adjustment history based on the acquired evaluation index data and the generation time of the evaluation index data (step S346).
- control parameter generation system 1C proceeds to the process in step S347.
- control parameter generation system 1C having the above configuration is configured such that even if it is necessary to re-adjust the control parameters, the vibration indicated by the evaluation index data exceeds a predetermined threshold. Control parameters are re-updated based on already generated evaluation index data other than evaluation index data newly generated in at least one time period. By doing so, it is possible to shorten the period for re-adjusting the control parameters compared to the case where the control parameters are re-adjusted from the beginning.
- the generation device 10 was described as being realized by one computer device. However, the generation device 10 does not necessarily need to be implemented by one computer device as long as it can implement similar functions.
- the generation device 10 may be realized by, for example, a plurality of computer devices that can communicate with each other.
- FIG. 26 is a block diagram showing an example of the configuration of the generation device 10 in a case where the generation device 10 is realized by a plurality of computer devices that can communicate with each other.
- the generation device 10 includes (1) an input section 11, an output section 12, an operation reception section 15, an image generation section 16, a display section 17, and a communication section 18. (2) a second computer device 110 including a control circuit 13, a machine learning model 14, and a communication unit 19; and (3) a communication unit 18 and a communication unit 19 are communicably connected. It may also be realized by a network 120 that supports
- the first computer device 100 may be, for example, a personal computer (personal computer), and the second computer device 110 may be a server.
- a personal computer personal computer
- the second computer device 110 may be a server.
- control parameter generation system 1 has (i) a tolerance that allows the sensor 30 to measure the position of the driven object 24 and evaluate that the driven object 24 has reached a predetermined target position; The description has been made assuming that the configuration is such that the measurement data representing the position of the driven object 24 after the time when it reaches the position is output, and (ii) the input unit 11 acquires the measurement data output from the sensor 30.
- control parameter generation system 1 is an example, and the control parameter generation system 1 is configured such that the input unit 11 determines that the drive object 24 has reached a permissible position after the time when the driven object 24 has reached a permissible position where it can be evaluated that the driven object 24 has reached a predetermined target position.
- the configuration is not necessarily limited to the above configuration as long as measurement data representing the position of the driven object 24 can be acquired.
- FIG. 27 is a block diagram showing the configuration of a control parameter generation system 1D that is an example of another configuration of the control parameter generation system 1 and does not include the sensor 30.
- the servo motor 23 has a built-in sensor that detects the motor position (for example, if the servo motor 23 is a rotary motor, the rotational position of the servo motor 23). Then, the servo motor 23 outputs an encoded value representing the motor position of the servo motor 23 detected by the built-in sensor. That is, the production device 20D outputs an encoded value.
- Such a sensor that detects the position of the motor is generally also referred to as an encoder.
- servo motors usually have a built-in encoder.
- the position of the servo motor 23 represented by this encoded value has a one-to-one correspondence with the position of the driven object 24. Therefore, the encoded value output by the servo motor 23 represents the position of the servo motor 23 and also represents the position of the driven object 24. Therefore, it can be said that the sensor built into the servo motor 23 also detects the position of the driven object 24 and outputs an encoded value representing the position of the driven object 24.
- the input unit 11 acquires the encoded value output from the servo motor 23 as measurement data.
- the general or specific aspects of the present disclosure may be realized in a system, device, method, integrated circuit, program, or non-transitory recording medium such as a computer-readable CD-ROM. Further, the present invention may be realized by any combination of systems, devices, methods, integrated circuits, programs, and non-transitory recording media. For example, the present disclosure may be implemented as a program for causing a computer device to execute the processing performed by the generation device.
- the production devices 20 and 20D of the present disclosure may include the generation devices 10 and 10A to 10C.
- the production apparatuses 20 and 20D of the present disclosure may include the control parameter generation systems 1 and 1A to 1D.
- the display device 40 may be placed separately from the generating devices 10, 10A to 10C, or may be placed independently of the generating devices 10, 10A to 10C. Further, the display device 40 may be incorporated into the production equipment 20, 20D.
- the sensor 30 and the sensors 30A and 30C with processing devices may be arranged separately from the production apparatuses 20 and 20D, or may be arranged independently of the production apparatuses 20 and 20D.
- the sensor 30 and the sensors with processing devices 30A and 30C may be installed and arranged in the generation devices 10 and 10A to 10C.
- the production devices 20 and 20D of the present disclosure are not limited to devices that produce products, but may include equipment, plants, and factories that produce products. Further, the present disclosure can be applied to a conveyance device in addition to the production devices 20 and 20D. Furthermore, the control parameter generation systems 1, 1A to 1D of the present disclosure can also be applied to equipment, equipment, plants, warehouses, or factories that produce products.
- the present disclosure can be widely used in systems that generate control parameters, etc.
- Control parameter generation system 10 10A, 10B, 10C Generation device 11, 11A, 11C Input section 12, 12B Output section 13, 13A, 13B, 13C Control circuit 14 Machine learning model 15 Operation reception Section 16 Image generation section 17 Display section 18, 19 Communication section 20, 20D Production device 21 Memory 22 Control circuit 23 Servo motor 24 Drive target 30 Sensor 30A, 30C Sensor with processing device 40 Display device 100 First computer device 110 No. 2 computer equipment 120 network
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Power Engineering (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Feedback Control In General (AREA)
Abstract
This method for generating a control parameter which is suitable for use in a generation device involves obtaining measurement data expressing location from a sensor for measuring the location of a drive target. Evaluation index data which expresses the vibrations of a drive target after the time at which the drive target reaches a permitted location is generated on the basis of the measurement data. The control parameter is updated, on the basis of the evaluation index data, by using a machine learning model for learning the relationship between the evaluation index data and the control parameter. Time data, which expresses one or more time periods during which the vibrations exhibited by the evaluation index data exceed a prescribed threshold, is outputted to a display device. When updating the updated control parameter again, the updated control parameter is updated again using the machine learning model on the basis of the already generated evaluation index data for data other than the evaluation index data which was newly generated during one or more time periods among said one or more time periods, and said newly generated evaluation index data.
Description
本開示は、制御パラメータの生成方法に関し、特に、生産装置に用いられる制御パラメータの生成方法に関する。
The present disclosure relates to a method of generating control parameters, and particularly to a method of generating control parameters used in production equipment.
従来、装置を制御するためのパラメータすなわち制御パラメータの生成方法が提案されている(例えば、特許文献1参照)。
Conventionally, a method for generating parameters for controlling a device, that is, a control parameter has been proposed (for example, see Patent Document 1).
従来、適正な制御パラメータの生成が望まれている。
Conventionally, it has been desired to generate appropriate control parameters.
そこで、本開示は、適正な制御パラメータを生成することができる制御パラメータの生成方法等を提供することを目的とする。
Therefore, an object of the present disclosure is to provide a control parameter generation method etc. that can generate appropriate control parameters.
本開示の一態様に係る制御パラメータの生成方法は、制御パラメータの生成装置が生成し、生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する。前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる前記制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力する。前記制御パラメータの生成方法は、以下の方法を有する。すなわち、前記センサから出力される前記位置を表す測定データを取得する。前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成する。前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation method according to one aspect of the present disclosure includes generating a control parameter by a control parameter generation device, using the control parameter by a production device, using a sensor to check the state of the production device, and displaying at least a part of the control parameter by a display device. Generate control parameters. The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. have things and. The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position. The control parameter generation method includes the following method. That is, measurement data representing the position output from the sensor is acquired. Based on the measurement data, evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position is generated in association with a generation time of the evaluation index data. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters. The updated control parameters are output to the production equipment for storage in the memory. Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device. When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. The updated control parameters are re-updated using the machine learning model based on the evaluation index data already generated.
本開示の他の一態様に係るプログラムは、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムである。
A program according to another aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
本開示の他の一態様に係る記録媒体は、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体である。
A recording medium according to another aspect of the present disclosure is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
本開示の他の一態様に係る制御パラメータの生成装置は、生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置である。前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力する。前記制御パラメータの生成装置は、入力部と、第2の制御回路と、出力部と、を備える。前記入力部は、前記センサから出力される前記位置を表す測定データを取得する。前記第2の制御回路は、前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成し、前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。前記出力部は、更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation device according to another aspect of the present disclosure is a control parameter generation device that is used by a production device, a sensor confirms a state of the production device, and a display device generates a control parameter that displays at least a part of the control parameter. This is a parameter generation device. The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven. The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position. The control parameter generation device includes an input section, a second control circuit, and an output section. The input unit acquires measurement data representing the position output from the sensor. Based on the measurement data, the second control circuit converts evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position into a generation time of the evaluation index data. The control parameters are updated using a machine learning model that is generated in association with each other and learns the relationship between the evaluation index data and the control parameters based on the evaluation index data. The output unit outputs the updated control parameters to the production device for storage in the memory. The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device. When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
本開示の他の一態様に係る制御パラメータの生成方法は、制御パラメータの生成装置が生成し、生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する。前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力する。前記制御パラメータの生成方法は、以下の方法を有する。前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得する。前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
In a control parameter generation method according to another aspect of the present disclosure, a control parameter generation device generates the control parameter, a production device uses it, a sensor with a processing device checks the state of the production device, and a display device displays at least the state of the production device. Generate some control parameters to display. The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores control parameters used by the control circuit to control the servo motor, and a driven object driven by the servo motor. and. The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. The position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output. The control parameter generation method includes the following method. Evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, is acquired in association with a generation time of the evaluation index data. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters. The updated control parameters are output to the production equipment for storage in the memory. Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device. When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. The updated control parameters are re-updated using the machine learning model based on the evaluation index data already generated.
本開示の他の一態様に係るプログラムは、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムである。
A program according to another aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
本開示の他の一態様に係る記録媒体は、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体である。
A recording medium according to another aspect of the present disclosure is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
本開示の他の一態様に係る制御パラメータの生成装置は、生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置である。前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力する。前記制御パラメータの生成装置は、入力部と、第2の制御回路と、出力部と、を備える。前記入力部は、前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得する。前記第2の制御回路は、前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。前記出力部は、更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation device according to another aspect of the present disclosure is used by a production device, a sensor with a processing device checks the state of the production device, and a display device generates control parameters that display at least a part of the control parameters. This is a control parameter generation device. The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven. The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. The position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output. The control parameter generation device includes an input section, a second control circuit, and an output section. The input unit may cause evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, to correspond to a generation time of the evaluation index data. Attach and obtain. The second control circuit updates the control parameter based on the evaluation index data using a machine learning model that learns a relationship between the evaluation index data and the control parameter. The output unit outputs the updated control parameters to the production device for storage in the memory. The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device. When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
本開示の上記態様に係る制御パラメータの生成方法、プログラム、記録媒体、および、制御パラメータの生成装置によれば、適正な制御パラメータを生成することができる。
According to the control parameter generation method, program, recording medium, and control parameter generation device according to the above aspects of the present disclosure, appropriate control parameters can be generated.
(本開示の一態様を得るに至った経緯)
従来、駆動対象物を駆動するサーボモータを備える生産装置に用いる制御パラメータの調整は、人手によっている。しかしながら、このような生産装置に用いる制御パラメータは、80個に及ぶこともあり、実際に調整する必要があるパラメータは、30~60個に及ぶこともある。また、対象とする生産装置に実装する駆動対象物、用途等に応じて、調整対象とするパラメータが異なる。 (How one aspect of the present disclosure was obtained)
Conventionally, control parameters used in production equipment equipped with a servo motor that drives a driven object have been manually adjusted. However, the number of control parameters used in such production equipment may reach 80, and the number of parameters that actually need to be adjusted may range from 30 to 60. Furthermore, the parameters to be adjusted differ depending on the drive target to be installed in the target production equipment, the application, and the like.
従来、駆動対象物を駆動するサーボモータを備える生産装置に用いる制御パラメータの調整は、人手によっている。しかしながら、このような生産装置に用いる制御パラメータは、80個に及ぶこともあり、実際に調整する必要があるパラメータは、30~60個に及ぶこともある。また、対象とする生産装置に実装する駆動対象物、用途等に応じて、調整対象とするパラメータが異なる。 (How one aspect of the present disclosure was obtained)
Conventionally, control parameters used in production equipment equipped with a servo motor that drives a driven object have been manually adjusted. However, the number of control parameters used in such production equipment may reach 80, and the number of parameters that actually need to be adjusted may range from 30 to 60. Furthermore, the parameters to be adjusted differ depending on the drive target to be installed in the target production equipment, the application, and the like.
このような事情から、従来、制御パラメータの調整は、熟練技術者が個別に対応している。しかし、それでも、制御パラメータの調整に数日オーダの期間が必要になることがある。
Due to these circumstances, conventionally, adjustment of control parameters has been handled individually by skilled engineers. However, adjustment of the control parameters may still require a period on the order of several days.
このような制御パラメータの調整は、振動による異音の発生の抑制、または、一旦動作を止め次の動作に入る際に前の動作の振動が次の動作に与える悪影響の抑制等のために必要である。
Adjustment of such control parameters is necessary to suppress the generation of abnormal noise due to vibration, or to suppress the negative effect of vibration from the previous operation on the next operation when stopping an operation and starting the next operation. It is.
例えば、特許文献1では、機械学習モデルを用いてこのような制御パラメータを調整することが提案されている。しかし、熟練技術者は、異音を聴覚で確認し、振動を視覚、触覚で確認する等、人間の五感を含めて制御パラメータを調整している。そのため、どのようなデータを機械学習モデルに入力すれば効果的に制御パラメータを調整することができるかについては、別途検討が必要である。
For example, Patent Document 1 proposes adjusting such control parameters using a machine learning model. However, skilled engineers adjust control parameters using all five human senses, such as checking abnormal noises by hearing and checking vibrations by sight and touch. Therefore, separate consideration is required as to what kind of data should be input to the machine learning model to effectively adjust the control parameters.
このような背景の中、発明者らは、どのようなデータを機械学習モデルに入力すれば、効果的に制御パラメータを調整することができるかについて、鋭意、検討、実験を重ねた。その結果、発明者らは、駆動対象物が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物の振動を表す評価指標データと制御パラメータとの間に相関関係があることを見出した。そして、発明者らは、上記評価指標データを機械学習モデルに入力すれば効果的に制御パラメータを調整できるとの知見を得た。
Against this background, the inventors have diligently studied and conducted repeated experiments to find out what kind of data should be input into a machine learning model to effectively adjust control parameters. As a result, the inventors found that there is a correlation between the control parameters and the evaluation index data representing the vibration of the driven object after the time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached a predetermined target position. I discovered something. The inventors have also found that control parameters can be effectively adjusted by inputting the evaluation index data into a machine learning model.
発明者らは、この知見を基に、更に、検討実験を重ね、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる制御パラメータの生成方法に想到した。
Based on this knowledge, the inventors conducted further studies and experiments, and came up with a control parameter generation method that can generate appropriate control parameters in a relatively short time and without requiring the use of skilled engineers. did.
しかしながら、発明者らは、この制御パラメータの生成方法を用いて制御パラメータの調整を行った後であっても、再度制御パラメータの調整をやり直したい場合があることに気が付いた。すなわち、発明者らは、この制御パラメータの生成方法を用いて制御パラメータの調整を行った後であっても、振動による異音の発生等が依然として発生する場合があることに気が付いた。このような場合には、もう一度最初からこの制御パラメータの生成方法を用いて制御パラメータの調整をやり直す必要があった。そして、もう一度やり直すことで、振動による異音の発生等が抑制されることを確認した。
However, the inventors have noticed that even after adjusting the control parameters using this control parameter generation method, there are cases where it is desired to adjust the control parameters again. That is, the inventors have noticed that even after adjusting the control parameters using this control parameter generation method, abnormal noises due to vibrations may still occur. In such a case, it is necessary to adjust the control parameters again from the beginning using this control parameter generation method. By repeating the process again, it was confirmed that abnormal noises caused by vibrations were suppressed.
発明者らは、このような現象が発生する原因を調査したところ、以下の知見を得た。
The inventors investigated the cause of such a phenomenon and obtained the following findings.
すなわち、機械学習モデルを用いて制御パラメータを調整する場合、調整対象とする生産装置等を常時人が監視していないことがある。このため、制御パラメータを調整している期間、調整対象とする生産装置等の周辺を人が歩く、調整対象とする生産装置等が設置される工場等の周辺の道路を大型トラック等が走行する、あるいは、調整対象とする生産装置等が設置される部屋等の出入口のドアが開け閉めされる等による、調整対象とする生産装置等の振動が、制御パラメータの調整に悪影響を及ぼしている場合があることが判明した。
That is, when adjusting control parameters using a machine learning model, the production equipment etc. to be adjusted may not be constantly monitored by humans. For this reason, during the period when control parameters are being adjusted, people walk around the production equipment, etc. to be adjusted, and large trucks, etc., drive on roads around factories, etc. where the production equipment, etc. to be adjusted are installed. Or, if the vibration of the production equipment, etc. to be adjusted is having an adverse effect on the adjustment of control parameters, due to the opening and closing of the entrance door of the room where the production equipment, etc. to be adjusted is installed. It turns out that there is.
図28は、発明者らが上記知見を得るに至る基となった実測データの一例であって、制御パラメータを調整している期間において、生産装置が設置される工場の周辺の道路を大型トラックが走行する場合における、レーザ変位計により測定された生産装置の変位の実測データと、走行しない場合における、レーザ変位計により測定された生産装置の変位の実測データとを示す、生産装置の振動波形である。
Figure 28 is an example of the actual measurement data that served as the basis for the inventors to obtain the above knowledge. Vibration waveform of the production equipment showing actual measurement data of the displacement of the production equipment measured by a laser displacement meter when the production equipment is traveling and actual measurement data of the displacement of the production equipment measured by the laser displacement meter when the production equipment is not traveling. It is.
図28において、横軸は経過時間であって、縦軸は生産装置の変位の振幅である。
In FIG. 28, the horizontal axis is the elapsed time, and the vertical axis is the amplitude of the displacement of the production equipment.
図28に示すように、発明者らは、周辺の道路を大型トラックが走行しない場合には生産装置の振幅が閾値の範囲内に十分に収まっている期間であっても、周辺の道路を大型トラックが走行している場合には生産装置の振動が閾値の範囲内に収まっていないという事実に気がついた。そして、発明者らは、制御パラメータを調整している期間において、周辺の道路を大型トラックが走行することに起因して生産装置の振動が閾値の範囲内に収まらないという上記現象が起こった場合には、制御パラメータの調整を行った後であっても、振動による異音の発生等が依然として発生してしまうことがあるという事実に気が付いた。
As shown in FIG. 28, the inventors discovered that even during a period when the amplitude of the production equipment is well within the threshold value when no large-sized trucks drive on the surrounding roads, the surrounding roads are It was noticed that the vibration of the production equipment was not within the threshold when the truck was running. The inventors discovered that when the above-mentioned phenomenon in which the vibration of the production equipment does not fall within the threshold range due to large trucks driving on the surrounding roads occurs during the period when the control parameters are being adjusted. It has been noticed that even after adjusting the control parameters, abnormal noises due to vibrations may still occur.
発明者らは、さらなる実測データの取得および解析を行い、生産装置の周辺を人が歩くことに起因して生産装置の振動が閾値の範囲内に収まらないという現象、生産装置が設置されている部屋の出入口のドアが開け閉めされることに起因して生産装置の振動が閾値の範囲内に収まらない現象が起こることを確認した。そして、制御パラメータを調整している期間において、このような現象が起こった場合にも、制御パラメータの調整を行った後であっても、振動による異音の発生等が依然として発生してしまうことがあるという事実を突き止めた。
The inventors acquired and analyzed further measured data, and discovered that the vibration of production equipment does not fall within the threshold range due to people walking around the production equipment, and that the production equipment is not installed. It was confirmed that the vibration of the production equipment did not fall within the threshold range due to the opening and closing of the door at the entrance to the room. Furthermore, even if such a phenomenon occurs during the period when the control parameters are being adjusted, abnormal noises due to vibration may still occur even after the control parameters have been adjusted. I found out that there is.
そこで、発明者らは、この新たに得られた知見を基に、更に、検討実験を重ね、下記本開示に係る制御パラメータの生成方法、プログラム、記録媒体、および、制御パラメータの生成装置に想到した。
Therefore, based on this newly obtained knowledge, the inventors conducted further studies and experiments, and came up with the following control parameter generation method, program, recording medium, and control parameter generation device according to the present disclosure. did.
本開示の一態様に係る制御パラメータの生成方法は、制御パラメータの生成装置が生成し、生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する。前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる前記制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力する。前記制御パラメータの生成方法は、以下の方法を有する。すなわち、前記センサから出力される前記位置を表す測定データを取得する。前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成する。前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記生成システムにおける表示装置に出力する。前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation method according to one aspect of the present disclosure includes generating a control parameter by a control parameter generation device, using the control parameter by a production device, using a sensor to check the state of the production device, and displaying at least a part of the control parameter by a display device. Generate control parameters. The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. have things and. The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position. The control parameter generation method includes the following method. That is, measurement data representing the position output from the sensor is acquired. Based on the measurement data, evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position is generated in association with a generation time of the evaluation index data. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters. The updated control parameters are output to the production equipment for storage in the memory. Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to a display device in the generation system. When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. The updated control parameters are re-updated using the machine learning model based on the evaluation index data already generated.
上記制御パラメータの生成方法によると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the control parameter generation method described above, by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate Control parameters can be generated.
さらには、上記制御パラメータの生成方法によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the control parameter generation method described above, even if it is necessary to re-adjust the control parameters, the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold. By re-updating the control parameters based on the evaluation index data that has already been generated, the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
また、前記機械学習モデルは、ベイズ最適化アルゴリズムを利用して、前記評価指標データと前記制御パラメータとの関係を学習するとしてもよい。
Furthermore, the machine learning model may learn the relationship between the evaluation index data and the control parameter using a Bayesian optimization algorithm.
機械学習モデルを、ベイズ最適化アルゴリズムを利用して、評価指標データと制御パラメータとの関係を学習するものとすることができる。
The machine learning model can use a Bayesian optimization algorithm to learn the relationship between evaluation index data and control parameters.
また、前記生産装置は、実装装置、加工装置、または、工作装置のいずれか1つを含むとしてもよい。
Furthermore, the production device may include any one of a mounting device, a processing device, or a machining device.
これにより、実装装置、加工装置、または、工作装置のいずれか1つである生産装置を対象として、適正な制御パラメータを生成することができる。
Thereby, appropriate control parameters can be generated for a production device that is any one of a mounting device, a processing device, or a machining device.
本開示の一態様に係るプログラムは、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムである。
A program according to one aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
上記プログラムによると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above program, by updating the control parameters based on evaluation index data indicating the vibration after the time when the permissible position is reached, appropriate control parameters are generated without the need of skilled engineers and in a relatively short time. can do.
さらには、上記プログラムによると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the above program, even if it is necessary to re-adjust the control parameters, the newly generated evaluation index data is generated during at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold. For other than the above, by re-updating the control parameters based on already generated evaluation index data, the period for re-adjusting the control parameters can be shortened compared to the case where the control parameters are re-adjusted from the beginning.
本開示の一態様に係る記録媒体は、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体である。
A recording medium according to one aspect of the present disclosure is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
上記記録媒体によると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above recording medium, by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate control parameters can be determined without the need of skilled engineers and in a relatively short time. can be generated.
さらには、上記記録媒体によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the recording medium, even if it is necessary to re-adjust the control parameters, the newly generated evaluation index is generated in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold. For items other than data, by re-updating the control parameters based on already generated evaluation index data, the period of re-adjustment can be shortened compared to the case of re-adjusting the control parameters from the beginning.
本開示の一態様に係る制御パラメータの生成装置は、生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置である。前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記制御パラメータの生成装置は、さらに入力部と、第2の制御回路と、出力部と、を備える。前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力する。前記入力部は、前記センサから出力される前記位置を表す測定データを取得する。前記第2の制御回路は、前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成し、前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。前記出力部は、更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation device according to one aspect of the present disclosure is a control parameter generation device that is used by a production device, a sensor confirms a state of the production device, and a display device generates a control parameter, at least a part of which is displayed. It is a generating device. The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven. The control parameter generation device further includes an input section, a second control circuit, and an output section. The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. Measures the position of an object and outputs measurement data representing the position. The input unit acquires measurement data representing the position output from the sensor. Based on the measurement data, the second control circuit converts evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position into a generation time of the evaluation index data. The control parameters are updated using a machine learning model that is generated in association with each other and learns the relationship between the evaluation index data and the control parameters based on the evaluation index data. The output unit outputs the updated control parameters to the production device for storage in the memory. The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device. When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
上記制御パラメータの生成装置によると、駆動対象物が許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above-mentioned control parameter generation device, the control parameters can be updated based on the evaluation index data indicating the vibration after the time when the driven object reaches the permissible position. , it is possible to generate appropriate control parameters.
さらには、上記制御パラメータの生成装置によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the control parameter generation device, even if it is necessary to re-adjust the control parameters, the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold. By re-updating the control parameters based on the evaluation index data that has already been generated, the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
本開示の他の一態様に係る制御パラメータの生成方法は、制御パラメータの生成装置が生成し、生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する。前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる前記制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、処理装置付きセンサと、を備える。前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力する。前記制御パラメータの生成方法は、以下の方法を有する。すなわち、前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得する。前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記生成システムにおける前記表示装置に出力する。前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
In a control parameter generation method according to another aspect of the present disclosure, a control parameter generation device generates the control parameter, a production device uses it, a sensor with a processing device checks the state of the production device, and a display device displays at least the state of the production device. Generate some control parameters to display. The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. and a sensor with a processing device. The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. The position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output. The control parameter generation method includes the following method. That is, evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position, which is output from the sensor with the processing device, is obtained in association with the generation time of the evaluation index data. do. Based on the evaluation index data, the control parameters are updated using a machine learning model that learns the relationship between the evaluation index data and the control parameters. The updated control parameters are output to the production equipment for storage in the memory. Time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold is output to the display device in the generation system. When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. The updated control parameters are re-updated using the machine learning model based on the evaluation index data already generated.
上記制御パラメータの生成方法によると、駆動対象物が許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the control parameter generation method described above, by updating the control parameters based on the evaluation index data indicating the vibration after the time when the driven object reaches the permissible position, the , it is possible to generate appropriate control parameters.
さらには、上記制御パラメータの生成方法によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the control parameter generation method described above, even if it is necessary to re-adjust the control parameters, the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold. By re-updating the control parameters based on the evaluation index data that has already been generated, the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
また、前記機械学習モデルは、ベイズ最適化アルゴリズムを利用して、前記評価指標データと前記制御パラメータとの関係を学習するとしてもよい。
Furthermore, the machine learning model may learn the relationship between the evaluation index data and the control parameter using a Bayesian optimization algorithm.
機械学習モデルを、ベイズ最適化アルゴリズムを利用して、評価指標データと制御パラメータとの関係を学習するものとすることができる。
The machine learning model can use a Bayesian optimization algorithm to learn the relationship between evaluation index data and control parameters.
また、前記評価指標データは、前記処理装置付きセンサにおいて、前記処理装置付きセンサにて測定された前記駆動対象物の位置に基づいて生成されるとしてもよい。
Furthermore, the evaluation index data may be generated in the sensor with a processing device based on the position of the driven object measured by the sensor with a processing device.
これにより、評価指標データを、処理装置付きセンサにより測定された駆動対象物の位置に基づいて生成されたものとすることができる。
Thereby, the evaluation index data can be generated based on the position of the driven object measured by the sensor with the processing device.
また、前記処理装置付きセンサは、画像処理装置を含むとしてもよい。
Furthermore, the sensor with a processing device may include an image processing device.
これにより、評価指標データを、画素処理に基づいて生成されたものとすることができる。
Thereby, the evaluation index data can be generated based on pixel processing.
また、前記生産装置は、実装装置、加工装置、または、工作装置のいずれか1つを含むとしてもよい。
Furthermore, the production device may include any one of a mounting device, a processing device, or a machining device.
これにより、実装装置、加工装置、または、工作装置のいずれか1つである生産装置を対象として、適正な制御パラメータを生成することができる。
Thereby, appropriate control parameters can be generated for a production device that is any one of a mounting device, a processing device, or a machining device.
本開示の一態様に係るプログラムは、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムである。
A program according to one aspect of the present disclosure is a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method.
上記プログラムによると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above program, by updating the control parameters based on evaluation index data indicating the vibration after the time when the permissible position is reached, appropriate control parameters are generated without the need of skilled engineers and in a relatively short time. can do.
さらには、上記プログラムによると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯を除外する時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the above program, even if it is necessary to re-adjust the control parameters, new vibrations are generated in a time period excluding at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold. By re-updating the control parameters based on the evaluation index data that has already been generated, the time required to re-adjust the control parameters can be shortened compared to when re-adjusting the control parameters from the beginning. can.
本開示の一態様に係る記録媒体は、上記制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体である。
A recording medium according to one aspect of the present disclosure is a recording medium on which a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method described above is recorded.
上記記録媒体によると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above recording medium, by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate control parameters can be determined without the need of skilled engineers and in a relatively short time. can be generated.
さらには、上記記録媒体によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the recording medium, even if it is necessary to re-adjust the control parameters, the newly generated evaluation index is generated in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold. For items other than data, by re-updating the control parameters based on already generated evaluation index data, the period of re-adjustment can be shortened compared to the case of re-adjusting the control parameters from the beginning.
本開示の他の一態様に係る制御パラメータの生成装置は、生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置である。前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備える。前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力する。前記制御パラメータの生成装置は、入力部と、第2の制御回路と、出力部と、を備える。前記入力部は、前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得する。前記第2の制御回路は、前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する。前記出力部は、更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する。前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力する。前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する。
A control parameter generation device according to another aspect of the present disclosure is used by a production device, a sensor with a processing device checks the state of the production device, and a display device generates control parameters that display at least a part of the control parameters. This is a control parameter generation device. The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven. The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. The position of the driven object is measured, and evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is output. The control parameter generation device includes an input section, a second control circuit, and an output section. The input unit may cause evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, to correspond to a generation time of the evaluation index data. Attach and obtain. The second control circuit updates the control parameter based on the evaluation index data using a machine learning model that learns a relationship between the evaluation index data and the control parameter. The output unit outputs the updated control parameters to the production device for storage in the memory. The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device. When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Other than the newly generated evaluation index data, the updated control parameters are re-updated using the machine learning model based on already generated evaluation index data.
上記制御パラメータの生成装置によると、許容位置に到達した時刻後の振動を示す評価指標データに基づき制御パラメータを更新することで、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
According to the above-mentioned control parameter generation device, by updating the control parameters based on the evaluation index data indicating the vibration after the time when the permissible position has been reached, appropriate Control parameters can be generated.
さらには、上記制御パラメータの生成装置によると、制御パラメータの調整をやり直す必要がある場合であっても、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については、既に生成された評価指標データに基づいて制御パラメータを再更新することで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。
Furthermore, according to the control parameter generation device, even if it is necessary to re-adjust the control parameters, the vibrations indicated by the evaluation index data are newly generated in at least one time period in which the vibrations exceed a predetermined threshold. By re-updating the control parameters based on the evaluation index data that has already been generated, the time required to re-adjust the control parameters can be shortened compared to the case of re-adjusting the control parameters from the beginning. .
以下、本開示の一態様に係る制御パラメータ生成システムの具体例について、図面を参照しながら説明する。ここで示す実施の形態は、いずれも本開示の一具体例を示すものである。従って、以下の実施の形態で示される数値、形状、構成要素、構成要素の配置および接続形態、ならびに、ステップ(工程)およびステップの順序等は、一例であって本開示を限定する趣旨ではない。また、各図は、模式図であり、必ずしも厳密に図示されたものではない。各図において、実質的に同一の構成に対しては同一の符号を付しており、重複する説明は省略または簡略化する。
Hereinafter, a specific example of a control parameter generation system according to one aspect of the present disclosure will be described with reference to the drawings. The embodiments shown here are all specific examples of the present disclosure. Therefore, the numerical values, shapes, components, arrangement and connection forms of the components, steps (processes) and order of steps, etc. shown in the following embodiments are merely examples and are not intended to limit the present disclosure. . Furthermore, each figure is a schematic diagram and is not necessarily strictly illustrated. In each figure, substantially the same configurations are denoted by the same reference numerals, and overlapping explanations will be omitted or simplified.
(実施の形態1)
以下、実施の形態1に係る制御パラメータ生成システム1について説明する。この制御パラメータ生成システム1は、駆動対象物24を駆動するサーボモータ23を備える生産装置20に用いる制御パラメータを生成するシステムである。 (Embodiment 1)
The controlparameter generation system 1 according to the first embodiment will be described below. This control parameter generation system 1 is a system that generates control parameters used in a production apparatus 20 that includes a servo motor 23 that drives a driven object 24 .
以下、実施の形態1に係る制御パラメータ生成システム1について説明する。この制御パラメータ生成システム1は、駆動対象物24を駆動するサーボモータ23を備える生産装置20に用いる制御パラメータを生成するシステムである。 (Embodiment 1)
The control
<構成>
図1は、実施の形態1に係る制御パラメータ生成システム1の概要を示す模式図である。 <Configuration>
FIG. 1 is a schematic diagram showing an overview of a controlparameter generation system 1 according to the first embodiment.
図1は、実施の形態1に係る制御パラメータ生成システム1の概要を示す模式図である。 <Configuration>
FIG. 1 is a schematic diagram showing an overview of a control
図2は、制御パラメータ生成システム1の構成を示すブロック図である。
FIG. 2 is a block diagram showing the configuration of the control parameter generation system 1.
図1に示すように、制御パラメータ生成システム1は、生成装置10と、生産装置20と、センサ30とを備える。センサ30は生産装置20とは別に備えてもよいが、生産装置20が備える構成要素としてよい。また。、センサ30は、生産装置20とは独立した検査装置またはその一部であってもよい。
As shown in FIG. 1, the control parameter generation system 1 includes a generation device 10, a production device 20, and a sensor 30. The sensor 30 may be provided separately from the production device 20, or may be a component included in the production device 20. Also. , the sensor 30 may be an inspection device independent of the production device 20 or a part thereof.
生産装置20は、機器を生産するために利用される装置であって、機器の加工、実装、搬送等を行う。生産装置20は、例えば、工場の生産ラインに設置される。生産装置20は、具体的には、例えば、LED(発光ダイオード)ボンダ、実装機、加工機、取り出しロボット等である。
The production device 20 is a device used to produce equipment, and processes, mounts, transports, etc. the equipment. The production device 20 is installed, for example, on a production line of a factory. Specifically, the production equipment 20 is, for example, an LED (light emitting diode) bonder, a mounting machine, a processing machine, a take-out robot, or the like.
図2に示すように、生産装置20は、メモリ21と、制御回路22と、サーボモータ23と、駆動対象物24とを備える。
As shown in FIG. 2, the production device 20 includes a memory 21, a control circuit 22, a servo motor 23, and a driven object 24.
サーボモータ23は、制御回路22によりその動作を制御され、駆動対象物24を駆動する。サーボモータ23は、例えば、回転式モータであってもよいし、リニアモータであってもよい。
The operation of the servo motor 23 is controlled by the control circuit 22 and drives the object 24 to be driven. The servo motor 23 may be, for example, a rotary motor or a linear motor.
駆動対象物24は、サーボモータ23により駆動される対象物である。駆動対象物24は、例えば、生産装置20において加工対象となる加工対象物、実装対象となる実装対象物、搬送対象となる搬送対象物等である。また、駆動対象物24は、例えば、生産装置20において取り扱われる対象物等を保持するアームであってもよく、これらの対象物等を搬送するトレーであってもよい。
The driven object 24 is an object driven by the servo motor 23. The driven object 24 is, for example, a processing object to be processed in the production apparatus 20, a mounting object to be mounted, a conveyance object to be transported, and the like. Furthermore, the driven object 24 may be, for example, an arm that holds objects handled in the production apparatus 20, or a tray that transports these objects.
制御回路22は、サーボモータ23に対して、駆動対象物24の位置を所定の目標位置にするための指令を出力することで、サーボモータ23を制御する。制御回路22がサーボモータ23に出力する指令は、例えば、サーボモータ23または駆動対象物24の位置を指令する位置指令であってもよいし、例えば、サーボモータ23のトルクを指令するトルク指令であってもよい。
The control circuit 22 controls the servo motor 23 by outputting a command to the servo motor 23 to position the driven object 24 at a predetermined target position. The command that the control circuit 22 outputs to the servo motor 23 may be, for example, a position command that commands the position of the servo motor 23 or the driven object 24, or may be, for example, a torque command that commands the torque of the servo motor 23. There may be.
制御回路22は、メモリ21に記憶される制御パラメータに基づいて、サーボモータ23を制御する。言い換えると、制御回路22は、サーボモータ23を制御する際に、メモリ21に記憶される制御パラメータを用いる。
The control circuit 22 controls the servo motor 23 based on the control parameters stored in the memory 21. In other words, the control circuit 22 uses the control parameters stored in the memory 21 when controlling the servo motor 23.
メモリ21は、制御回路22がサーボモータ23を制御する際に用いる制御パラメータを記憶する。メモリ21が記憶する制御パラメータは、生成装置10から出力された制御パラメータである。図3は、メモリ21が記憶する制御パラメータの一例を示す模式図である。
The memory 21 stores control parameters used by the control circuit 22 to control the servo motor 23. The control parameters stored in the memory 21 are the control parameters output from the generation device 10. FIG. 3 is a schematic diagram showing an example of control parameters stored in the memory 21.
図3に示すように、メモリ21が記憶する制御パラメータには、例えば、駆動対象物24の振動周波数を調整するパラメータa1、a2、駆動対象物24の高速化を調整するパラメータb1、b2、駆動対象物24の振動特性における特異点の深さを調整するパラメータc1、c2、駆動対象物24の振動振幅を調整するパラメータd1、d2等が含まれる。
As shown in FIG. 3, the control parameters stored in the memory 21 include, for example, parameters a1 and a2 for adjusting the vibration frequency of the driven object 24, parameters b1 and b2 for adjusting the speed increase of the driven object 24, and parameters b1 and b2 for adjusting the speed of the driven object 24. The parameters include parameters c1 and c2 for adjusting the depth of a singular point in the vibration characteristics of the object 24, parameters d1 and d2 for adjusting the vibration amplitude of the driven object 24, and the like.
一般に、制御パラメータの中には、例えば、高速化を調整するパラメータb1、b2と、振動特性における特異点の深さを調整するパラメータc1、c2、および、振動振幅を調整するパラメータd1、d2とのように、互いにトレードオフの関係性があるパラメータが含まれる。
In general, the control parameters include, for example, parameters b1 and b2 that adjust speed, parameters c1 and c2 that adjust the depth of singularity in vibration characteristics, and parameters d1 and d2 that adjust vibration amplitude. This includes parameters that have a trade-off relationship with each other, such as.
再び、図1および図2に戻って、制御パラメータ生成システム1の説明を続ける。
Returning again to FIGS. 1 and 2, the description of the control parameter generation system 1 will be continued.
センサ30は、駆動対象物24の位置を測定する。そして、センサ30は、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを、生成装置10に出力する。
The sensor 30 measures the position of the driven object 24. Then, the sensor 30 outputs to the generation device 10 measurement data representing the position of the driven object 24 after the time when the driven object 24 reaches an allowable position where it can be evaluated that the driven object 24 has reached a predetermined target position.
図4は、生産装置20が、駆動対象物24を目標位置に駆動する際における、目標位置に対する駆動対象物24の位置偏差の推移の一例を示す模式図である。
FIG. 4 is a schematic diagram showing an example of the transition of the positional deviation of the driven object 24 with respect to the target position when the production apparatus 20 drives the driven object 24 to the target position.
図4において、横軸は時刻を示し、縦軸は、目標位置に対する駆動対象物24の位置偏差を示す。
In FIG. 4, the horizontal axis indicates time, and the vertical axis indicates the positional deviation of the driven object 24 with respect to the target position.
図4に示すように、本明細書において、許容位置とは、目標位置からの位置偏差が要求精度以内の位置のことをいう。また、図4に示すように、本明細書において、目標位置に到達したと評価できる許容位置に到達した時刻(以下、「整定時刻」とも称する。)とは、駆動対象物24が許容位置に到達した後において再び許容位置から逸脱しなくなった場合における、最後に許容位置に到達した時刻(図4中の時刻sett_t)のことをいう。
As shown in FIG. 4, in this specification, an allowable position refers to a position where the positional deviation from the target position is within the required accuracy. Further, as shown in FIG. 4, in this specification, the time when the driven object 24 reaches the permissible position (hereinafter also referred to as "settling time") at which it can be evaluated that the target position has been reached is the time when the driven object 24 reaches the permissible position. This refers to the time when the permissible position is finally reached (time sett_t in FIG. 4) when the permissible position does not deviate again after reaching the permissible position.
図5は、センサ30が出力する測定データの一例を示すデータ構成図である。
FIG. 5 is a data configuration diagram showing an example of measurement data output by the sensor 30.
図5に示すように、測定データは、一例として、許容位置に到達した時刻後の経過時間[ms]と、目標位置からの偏差量[mm]とを一対一に対応付けたデータである。
As shown in FIG. 5, the measurement data is, for example, data in which the elapsed time [ms] after the time when the permissible position is reached and the deviation amount [mm] from the target position are associated in one-to-one correspondence.
再び、図1および図2に戻って、制御パラメータ生成システム1の説明を続ける。
Returning again to FIGS. 1 and 2, the description of the control parameter generation system 1 will be continued.
生成装置10は、メモリ21が記憶している制御パラメータ(以下、「記憶中制御パラメータ」とも称する)を更新するための制御パラメータ(以下、「更新用制御パラメータ」とも称する)を生成する。より具体的には、生成装置10は、生産装置20が記憶中制御パラメータを用いて駆動対象物24を目標位置に駆動した結果、センサ30から出力される測定データを逐次取得し、取得した測定データに基づいて更新用制御パラメータを逐次生成し、生成した更新用制御パラメータを生産装置20に逐次出力する。
The generation device 10 generates control parameters (hereinafter also referred to as "updating control parameters") for updating the control parameters stored in the memory 21 (hereinafter also referred to as "stored control parameters"). More specifically, the generation device 10 sequentially acquires measurement data output from the sensor 30 as a result of the production device 20 driving the driven object 24 to the target position using the stored control parameters, and uses the acquired measurement data. Update control parameters are sequentially generated based on the data, and the generated update control parameters are sequentially output to the production apparatus 20.
図2に示すように、生成装置10は、入力部11と、出力部12と、制御回路13と、機械学習モデル14と、操作受け付け部15と、画像生成部16と、表示部17とを備える。
As shown in FIG. 2, the generation device 10 includes an input section 11, an output section 12, a control circuit 13, a machine learning model 14, an operation reception section 15, an image generation section 16, and a display section 17. Be prepared.
生成装置10は、例えば、プロセッサと、メモリと、入出力インターフェースとを備えるコンピュータ装置において、プロセッサがメモリに記憶されるプログラムを実行することで実現される。このようなコンピュータ装置は、例えば、パーソナルコンピュータ(パソコン)である。
The generation device 10 is realized, for example, in a computer device including a processor, a memory, and an input/output interface, when the processor executes a program stored in the memory. Such a computer device is, for example, a personal computer.
入力部11は、センサ30から出力された測定データを取得する。入力部11は、例えば、信号を入力する電子部品または電子回路であってよい。
The input unit 11 acquires the measurement data output from the sensor 30. The input unit 11 may be, for example, an electronic component or an electronic circuit that inputs a signal.
制御回路13は、入力部11により取得された測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを生成し、生成した評価指標データに基づいて、評価指標データと制御パラメータとの関係を学習する機械学習モデル14を用いて、制御パラメータを更新する。
The control circuit 13 generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data acquired by the input unit 11, and calculates the evaluation index data based on the generated evaluation index data. The control parameters are updated using the machine learning model 14 that learns the relationship with the control parameters.
制御回路13が生成する評価指標データは、以下の(1)~(3)のうちのいずれか1つ、または(4)である。すなわち、
(1)整定時刻後の駆動対象物24の振動を示す偏差波形と基準軸とで囲まれる面積の総和(以下、「総和面積評価指標データ」とも称する)。
(2)整定時刻後の駆動対象物24の振動を示す偏差波形のバラツキ度合い(以下、「バラツキ度合い評価指標データ」とも称する)。
(3)整定時刻後の駆動対象物24の振動を示す偏差波形の実効値(以下、「実効値評価指標データ」とも称する)。
のうちのいずれか1つ、または、
(4)駆動対象物24が駆動され始めた時刻から整定時刻までの時間(以下、「整定時間評価指標データ」とも称する)。
である。 The evaluation index data generated by thecontrol circuit 13 is any one of the following (1) to (3) or (4). That is,
(1) The sum total of the area surrounded by the deviation waveform indicating the vibration of the drivenobject 24 after the settling time and the reference axis (hereinafter also referred to as "total area evaluation index data").
(2) Degree of variation in the deviation waveform indicating the vibration of the drivenobject 24 after the settling time (hereinafter also referred to as "dispersion degree evaluation index data").
(3) Effective value of the deviation waveform indicating the vibration of the drivenobject 24 after the settling time (hereinafter also referred to as "effective value evaluation index data").
any one of the following, or
(4) Time from the time when the drivenobject 24 starts to be driven to the settling time (hereinafter also referred to as "settling time evaluation index data").
It is.
(1)整定時刻後の駆動対象物24の振動を示す偏差波形と基準軸とで囲まれる面積の総和(以下、「総和面積評価指標データ」とも称する)。
(2)整定時刻後の駆動対象物24の振動を示す偏差波形のバラツキ度合い(以下、「バラツキ度合い評価指標データ」とも称する)。
(3)整定時刻後の駆動対象物24の振動を示す偏差波形の実効値(以下、「実効値評価指標データ」とも称する)。
のうちのいずれか1つ、または、
(4)駆動対象物24が駆動され始めた時刻から整定時刻までの時間(以下、「整定時間評価指標データ」とも称する)。
である。 The evaluation index data generated by the
(1) The sum total of the area surrounded by the deviation waveform indicating the vibration of the driven
(2) Degree of variation in the deviation waveform indicating the vibration of the driven
(3) Effective value of the deviation waveform indicating the vibration of the driven
any one of the following, or
(4) Time from the time when the driven
It is.
制御回路13が生成する評価指標データは、制御パラメータ生成システム1を利用するユーザにより指定される。
The evaluation index data generated by the control circuit 13 is specified by the user using the control parameter generation system 1.
以下、制御回路13が生成する、総和面積評価指標データとバラツキ度合い評価指標データと実効値評価指標データとについて、図4を参照しながらより詳細に説明する。
Hereinafter, the total area evaluation index data, variation degree evaluation index data, and effective value evaluation index data generated by the control circuit 13 will be explained in more detail with reference to FIG. 4.
図4に示すように、制御回路13が生成する総和面積データは、より具体的には、目標位置に対する駆動対象物の位置偏差fを時刻tの変数として、f=E(t)で表した場合に、(式1)、(式2)、または、(式3)に示される数式により算出される。
As shown in FIG. 4, the total area data generated by the control circuit 13 is more specifically expressed as f=E(t), where the positional deviation f of the driven object with respect to the target position is taken as a variable at time t. In this case, it is calculated by the formula shown in (Formula 1), (Formula 2), or (Formula 3).
ここで、積分区間の始点sett_tは、整定時刻であって、積分区間の終点cal_eはユーザにより設定された時刻である。なお、ここでは、積分区間の始点sett_tが整定時刻であるとして説明するが、積分区間の始点sett_tは、整定時刻であってもよいし、整定時刻後の任意の時刻であってもよい。
Here, the starting point sett_t of the integral interval is the settling time, and the end point cal_e of the integral interval is the time set by the user. Note that although the explanation will be given here assuming that the starting point sett_t of the integral interval is the settling time, the starting point sett_t of the integral interval may be the settling time or any time after the settling time.
cal_eは、例えば、目標位置に対する駆動対象物24の位置偏差が要求精度の1/10以内に到達した後において再び位置偏差が要求精度の1/10以内から逸脱しなくなった場合における、最後に位置偏差が要求精度の1/10以内に到達した時刻に設定されてもよいし、例えば、入力部11が取得した測定データに含まれるデータ数が所定の数に達する時刻に設定されてもよいし、整定時刻から指定された時間経過した時刻に設定されてもよい。
cal_e is, for example, the final position when the positional deviation of the driven object 24 with respect to the target position reaches within 1/10 of the required accuracy and then the positional deviation no longer deviates from within 1/10 of the required accuracy. It may be set at the time when the deviation reaches within 1/10 of the required accuracy, or for example, it may be set at the time when the number of data included in the measurement data acquired by the input unit 11 reaches a predetermined number. , it may be set to a time when a specified time has elapsed from the settling time.
制御回路13が生成するバラツキ度合い評価指標データは、より具体的には、def = cal_e - sett_tとした場合に、(式4)に示される数式により算出される。
More specifically, the degree of variation evaluation index data generated by the control circuit 13 is calculated by the formula shown in (Formula 4) when def = cal_e - sett_t.
制御回路13が生成する実効値評価指標データは、より具体的には、(式5)に示される数式により算出される。
More specifically, the effective value evaluation index data generated by the control circuit 13 is calculated by the formula shown in (Formula 5).
再び、図1および図2に戻って、制御パラメータ生成システム1の説明を続ける。
Returning again to FIGS. 1 and 2, the description of the control parameter generation system 1 will be continued.
機械学習モデル14は、評価指標データと制御パラメータとの関係を学習する機械学習モデルであって、制御回路13から評価指標データが逐次入力される毎に、評価指標データと制御パラメータとの関係を学習し、学習結果に基づいて制御回路13に逐次制御パラメータを出力する。
The machine learning model 14 is a machine learning model that learns the relationship between evaluation index data and control parameters, and learns the relationship between evaluation index data and control parameters each time evaluation index data is sequentially input from the control circuit 13. It learns and sequentially outputs control parameters to the control circuit 13 based on the learning results.
機械学習モデル14は、より具体的には、ベイズ最適化アルゴリズムを利用して、制御回路13から評価指標データが逐次入力される毎に、評価指標データと、その評価指標データに対応する制御パラメータ(以下、「対応制御パラメータ」とも称する)との関係を学習して、学習結果に基づいて、最も評価指標データが小さくなると予測される制御パラメータを、制御回路13に出力する。
More specifically, the machine learning model 14 uses a Bayesian optimization algorithm to calculate evaluation index data and control parameters corresponding to the evaluation index data each time the evaluation index data is sequentially input from the control circuit 13. (hereinafter also referred to as "corresponding control parameters"), and outputs to the control circuit 13 the control parameters predicted to have the smallest evaluation index data based on the learning results.
図6は、機械学習モデル14が、評価指標データと対応制御パラメータとの関係を学習し、学習した結果に基づいて、最も評価指標データが小さくなると予測される制御パラメータを出力する様子を示す模式図である。
FIG. 6 is a schematic diagram showing how the machine learning model 14 learns the relationship between evaluation index data and corresponding control parameters, and outputs the control parameters that are predicted to cause the smallest evaluation index data based on the learned results. It is a diagram.
図6において、横軸は、制御パラメータの値であり、縦軸は、評価指標データの値である。また、図6において、丸でプロットされた点は、機械学習モデル14に過去に入力された評価指標データの値のそれぞれについての、その評価指標データの対応制御パラメータの値を示す点である。
In FIG. 6, the horizontal axis is the value of the control parameter, and the vertical axis is the value of the evaluation index data. In addition, in FIG. 6, the points plotted with circles indicate the values of the corresponding control parameters of the evaluation index data for each value of the evaluation index data input into the machine learning model 14 in the past.
図6に示すように、機械学習モデル14は、ベイズ最適化アルゴリズムを利用して、評価指標データと対応制御パラメータとの関係を、ある程度の範囲を持って予測し、予測した関係を用いて、最も評価指標データが小さくなると予測される制御パラメータを出力する。
As shown in FIG. 6, the machine learning model 14 uses a Bayesian optimization algorithm to predict the relationship between evaluation index data and corresponding control parameters within a certain range, and uses the predicted relationship to A control parameter that is predicted to have the smallest evaluation index data is output.
再び、図1および図2に戻って、制御パラメータ生成システム1の説明を続ける。
Returning again to FIGS. 1 and 2, the description of the control parameter generation system 1 will be continued.
出力部12は、制御回路13により更新された制御パラメータを、メモリ21に記憶させるために、生産装置20に出力する。出力部12は、例えば、信号を出力する電子部品または電子回路であってよい。
The output unit 12 outputs the control parameters updated by the control circuit 13 to the production device 20 in order to store them in the memory 21. The output unit 12 may be, for example, an electronic component or an electronic circuit that outputs a signal.
操作受け付け部15は、制御パラメータ生成システム1を利用するユーザによる、生成装置10への入力操作を受け付ける。操作受け付け部15は、例えば、ボタンやキーボードといった入力装置であってよい。
The operation accepting unit 15 accepts an input operation to the generating device 10 by a user using the control parameter generating system 1. The operation receiving unit 15 may be, for example, an input device such as a button or a keyboard.
表示部17は、制御パラメータ生成システム1を利用するユーザに提供する画像を表示する。表示部17は、例えば、液晶モニタ、有機EL(Electro Luminescence)モニタあるいはCRT(Cathode Ray Tube)モニタであってよい。
The display unit 17 displays images provided to the user who uses the control parameter generation system 1. The display unit 17 may be, for example, a liquid crystal monitor, an organic EL (Electro Luminescence) monitor, or a CRT (Cathode Ray Tube) monitor.
画像生成部16は、表示部17が表示する画像を生成する。
The image generation unit 16 generates an image to be displayed on the display unit 17.
図7は、画像生成部16が生成する画像の一例である。図7は、制御パラメータ生成システム1が、後述する第1の制御パラメータ調整処理を開始する際に、制御パラメータ生成システム1を利用するユーザに対して、制御回路13が生成する評価指標データをどの評価指標データとするかの指定を促す画像(以下、「評価指標データ指定画像」とも称する)の一例となっている。画像生成部16は、例えば、グラフィックボードであってよい。
FIG. 7 is an example of an image generated by the image generation unit 16. FIG. 7 shows how the control parameter generation system 1 sends evaluation index data generated by the control circuit 13 to a user who uses the control parameter generation system 1 when starting a first control parameter adjustment process to be described later. This is an example of an image (hereinafter also referred to as "evaluation index data designation image") that prompts the user to specify whether or not to use the evaluation index data. The image generation unit 16 may be, for example, a graphic board.
図7に示すように、生成装置10は、評価指標データ指定画像を表示部17に表示させることで、ユーザに対して、制御回路13が生成する評価指標データとして、整定時間評価指標データの指定、または、総和面積評価指標データと、バラツキ度合い評価指標データと、実効値評価指標データと(以下、これら3つの評価指標データのことを「振動評価指標データ」とも称する)のうちのいずれか1つの指定を促す。
As shown in FIG. 7, the generation device 10 displays an evaluation index data designation image on the display unit 17, thereby prompting the user to designate settling time evaluation index data as the evaluation index data generated by the control circuit 13. , or any one of total area evaluation index data, variation degree evaluation index data, and effective value evaluation index data (hereinafter, these three evaluation index data are also referred to as "vibration evaluation index data"). Encourage one person to specify one.
ユーザは、例えば、評価指標データ指定画像に対して、「整定時間」の横にチェックを入れることで、評価指標データとして、整定時間評価指標データを指定する。ユーザは、例えば、評価指標データ指定画像に対して、「振動」の横にチェックを入れ、かつ、「面積」の横に丸印を入れることで、評価指標データとして、面積総和評価指標データを指定する。ユーザは、例えば、評価指標データ指定画像に対して、「振動」の横にチェックを入れ、かつ、「分散」の横に丸印を入れることで、評価指標データとして、バラツキ度合い評価指標データを指定する。ユーザは、例えば、評価指標データ指定画像に対して、「振動」の横にチェックを入れ、かつ、「RMS(実効値)」の横に丸印を入れることで、評価指標データとして、実効値評価指標データを指定する。
The user specifies the settling time evaluation index data as the evaluation index data, for example, by placing a check next to "settling time" in the evaluation index data designation image. For example, the user can select area sum evaluation index data as evaluation index data by checking the box next to "vibration" and putting a circle next to "area" in the evaluation index data specification image. specify. For example, the user can select the degree of variation evaluation index data as evaluation index data by checking the box next to "vibration" and putting a circle next to "dispersion" in the evaluation index data specification image. specify. For example, the user can select the effective value as evaluation index data by checking the box next to "vibration" and putting a circle next to "RMS (effective value)" in the evaluation index data specification image. Specify evaluation index data.
また、ユーザは、例えば、評価指標データ指定画像に対して、「整定時間」の横と「振動」の横との双方にチェックを入れることで、評価指標データとして、整定時間評価指標データおよび振動評価指標データを指定する。この場合、制御回路13は、整定時間評価指標データの値と、振動評価指標データの値との和を評価指標データの値とする評価指標データとしてもよいし、整定時間評価指標データの値と、振動評価指標データの値とのそれぞれに対して、予め定められた重み付け係数を乗じた値の和を評価指標データの値とする評価指標データとしてもよい。
In addition, the user can, for example, check the boxes next to "Settling time" and "Vibration" in the evaluation index data specification image to select the settling time evaluation index data and vibration as evaluation index data. Specify evaluation index data. In this case, the control circuit 13 may set the value of the settling time evaluation index data to be the sum of the value of the settling time evaluation index data and the value of the vibration evaluation index data. , vibration evaluation index data, and the value of the evaluation index data may be the sum of the values obtained by multiplying each value by a predetermined weighting coefficient.
図8は、画像生成部16が生成する画像の一例である。図8は、制御パラメータ生成システム1が、後述する第1の制御パラメータ調整処理を実行している最中に、制御パラメータ生成システム1を利用するユーザに対して、第1の制御パラメータ調整処理における評価指標データの値の状況を提示する画像(以下、「状況提示画像」とも称する)の一例となっている。
FIG. 8 is an example of an image generated by the image generation unit 16. FIG. 8 shows that while the control parameter generation system 1 is executing the first control parameter adjustment process, which will be described later, a user using the control parameter generation system 1 is asked to perform the first control parameter adjustment process. This is an example of an image that presents the status of the value of evaluation index data (hereinafter also referred to as a "situation presentation image").
図8中のグラフにおいて、横軸は、生成装置10が逐次出力した制御パラメータの回数であり、縦軸は、生成装置10が逐次出力した制御パラメータのそれぞれに対応する評価指標データの値である。また、図8中のグラフにおいて、実線の折れ線は、生成装置10が出力した制御パラメータに対応する評価指標データの値の最小値の推移を示す。
In the graph in FIG. 8, the horizontal axis is the number of control parameters sequentially output by the generating device 10, and the vertical axis is the value of evaluation index data corresponding to each of the control parameters sequentially output by the generating device 10. . Furthermore, in the graph in FIG. 8, the solid line indicates the transition of the minimum value of the evaluation index data corresponding to the control parameter output by the generation device 10.
図8に示すように、状況提示画像の右上に、「中断」と記されたアイコン(以下、「中断アイコン」とも称する)と、「再開」と記されたアイコン(以下、「再開アイコン」とも称する)とが配置されている。
As shown in Figure 8, in the upper right corner of the status presentation image, there are two icons marked "Suspend" (hereinafter also referred to as "suspension icon") and an icon marked "Resume" (hereinafter also referred to as "resume icon"). ) are located.
中断アイコンは、例えば、ユーザが状況提示画像を視認した結果、生成装置10が出力した制御パラメータに対応する評価指標データの値が十分に小さくなっているために、これ以上、第1の制御パラメータ調整処理を継続させる必要がないと判断した場合に、中断アイコンをクリックすることで、第1の制御パラメータ調整処理を中断させることができるアイコンである。
The interruption icon indicates, for example, that the value of the evaluation index data corresponding to the control parameter output by the generation device 10 has become sufficiently small as a result of the user visually confirming the situation presentation image, so that the first control parameter is no longer used. This icon allows you to interrupt the first control parameter adjustment process by clicking the interrupt icon when it is determined that there is no need to continue the adjustment process.
再開アイコンは、例えば、ユーザが状況提示画像を視認した結果、生成装置10が出力した制御パラメータに対応する評価指標データの値が十分に小さくなっていないために、第1の制御パラメータ調整処理を継続させる必要があると判断した場合に、再開アイコンをクリックすることで、第1の制御パラメータ調整処理を継続させることができるアイコンである。
The restart icon indicates, for example, that the first control parameter adjustment process is performed because the value of the evaluation index data corresponding to the control parameter output by the generation device 10 has not become sufficiently small as a result of the user viewing the situation presentation image. This icon allows you to continue the first control parameter adjustment process by clicking the restart icon when it is determined that it is necessary to continue the process.
<動作>
以下、上記構成の制御パラメータ生成システム1が行う動作について説明する。 <Operation>
The operation performed by the controlparameter generation system 1 having the above configuration will be described below.
以下、上記構成の制御パラメータ生成システム1が行う動作について説明する。 <Operation>
The operation performed by the control
制御パラメータ生成システム1は、メモリ21が記憶する制御パラメータを適正な値に調整する第1の制御パラメータ調整処理を実行する。第1の制御パラメータ調整処理は、例えば、制御パラメータ生成システム1を利用するユーザが、生成装置10に対して、第1の制御パラメータ調整処理を開始する旨の操作を行うことで開始される。
The control parameter generation system 1 executes a first control parameter adjustment process that adjusts the control parameters stored in the memory 21 to appropriate values. The first control parameter adjustment process is started, for example, when a user using the control parameter generation system 1 performs an operation on the generation device 10 to start the first control parameter adjustment process.
図9は、第1の制御パラメータ調整処理のシーケンス図であり、図10は、第1の制御パラメータ調整処理のフローチャートである。
FIG. 9 is a sequence diagram of the first control parameter adjustment process, and FIG. 10 is a flowchart of the first control parameter adjustment process.
図9、図10に示すように、第1の制御パラメータ調整処理が開始されると、生成装置10は、第1の制御パラメータ調整処理を実行するための所定のプログラムを起動する(ステップS10)。
As shown in FIGS. 9 and 10, when the first control parameter adjustment process is started, the generation device 10 starts a predetermined program for executing the first control parameter adjustment process (step S10). .
所定のプログラムが起動されると、出力部12は、制御パラメータの初期値をメモリ21に出力する(ステップS15)。このとき、出力部12は、例えば、予め定められた値からなる制御パラメータの初期値を出力するとしてもよいし、例えば、ユーザにより指定された値からなる制御パラメータの初期値を出力するとしてもよいし、例えば、ユーザにより指定された算出方法により算出された値からなる制御パラメータの初期値を出力するとしてもよい。
When the predetermined program is started, the output unit 12 outputs the initial values of the control parameters to the memory 21 (step S15). At this time, the output unit 12 may, for example, output an initial value of the control parameter consisting of a predetermined value, or may output an initial value of the control parameter consisting of a value specified by the user. Alternatively, for example, the initial value of the control parameter may be output, which is a value calculated by a calculation method specified by the user.
出力部12から制御パラメータが出力されると、メモリ21は、出力部12から出力された制御パラメータを記憶する(ステップS20)。すると、制御回路22は、メモリ21が記憶する制御パラメータに基づいて、駆動対象物24の位置を所定の目標位置にするための指令を生成してサーボモータ23に出力する。これにより、制御回路22は、サーボモータ23を制御する(ステップS25)。すると、サーボモータ23は、制御回路22から出力された指令に基づいて、駆動対象物24を駆動する(ステップS30)。
When the control parameters are output from the output unit 12, the memory 21 stores the control parameters output from the output unit 12 (step S20). Then, the control circuit 22 generates a command to position the driven object 24 to a predetermined target position based on the control parameters stored in the memory 21 and outputs it to the servo motor 23. Thereby, the control circuit 22 controls the servo motor 23 (step S25). Then, the servo motor 23 drives the driven object 24 based on the command output from the control circuit 22 (step S30).
駆動対象物24がサーボモータ23により駆動されると、センサ30は、駆動対象物24の位置を測定し(ステップS35)、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを、入力部11に出力する。すると、入力部11は、センサ30から出力された測定データを取得する(ステップS40)。
When the driven object 24 is driven by the servo motor 23, the sensor 30 measures the position of the driven object 24 (step S35), and moves the driven object 24 to an allowable position where it can be evaluated that it has reached a predetermined target position. Measurement data representing the position of the driven object 24 after the time reached is output to the input unit 11 . Then, the input unit 11 acquires the measurement data output from the sensor 30 (step S40).
入力部11が測定データを取得すると、制御回路13は、測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを生成する(ステップS45)。そして、制御回路13は、評価指標データの生成が、所定の条件を満たしているか否かを調べる(ステップS50)。ここで、所定の条件とは、例えば、評価指標データの更新回数(すなわち生成回数)が所定の回数に到達したという条件、制御パラメータの調整開始から所定の時間が経過したという条件等である。ここで、所定の回数、所定の時間は、あらかじめ定められるとしていてもよいし、制御パラメータ生成システム1を利用するユーザにより指定されるとしてもよい。
When the input unit 11 acquires the measurement data, the control circuit 13 generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data (step S45). Then, the control circuit 13 checks whether the generation of evaluation index data satisfies a predetermined condition (step S50). Here, the predetermined conditions include, for example, the condition that the number of times the evaluation index data has been updated (that is, the number of times it has been generated) has reached a predetermined number, the condition that a predetermined time has elapsed since the start of adjustment of the control parameters, and the like. Here, the predetermined number of times and the predetermined time may be determined in advance, or may be specified by the user using the control parameter generation system 1.
ステップS50の処理において、評価指標データの生成が、所定の条件を満たしていない場合に(ステップS50:No)、制御回路13は、生成した評価指標データを、機械学習モデル14に出力する。すると、機械学習モデル14は、評価指標データと、対応制御パラメータとの関係を学習して(ステップS55)、最も評価指標データの値が小さくなると予測される制御パラメータを出力する。すると、制御回路13は、機械学習モデル14から新たに出力された制御パラメータで、前回、出力部12が出力した制御パラメータを更新する(ステップS60)。すると、出力部12は、制御回路13により新たに更新された制御パラメータを、メモリ21に送信する(ステップS65)。
In the process of step S50, if the generation of evaluation index data does not satisfy the predetermined condition (step S50: No), the control circuit 13 outputs the generated evaluation index data to the machine learning model 14. Then, the machine learning model 14 learns the relationship between the evaluation index data and the corresponding control parameter (step S55), and outputs the control parameter that is predicted to have the smallest value of the evaluation index data. Then, the control circuit 13 updates the control parameters previously output by the output unit 12 with the control parameters newly output from the machine learning model 14 (step S60). Then, the output unit 12 transmits the control parameters newly updated by the control circuit 13 to the memory 21 (step S65).
ステップS65の処理が終了すると、制御パラメータ生成システム1は、ステップS20の処理に進んで、ステップS20以下の処理を繰り返す。
When the process of step S65 is completed, the control parameter generation system 1 proceeds to the process of step S20, and repeats the process from step S20 onwards.
ステップS50の処理において、生成した評価指標データが、所定の条件を満たしている場合に(ステップS50:Yes)、制御パラメータ生成システム1は、その第1の制御パラメータ調整処理を終了する。
In the process of step S50, if the generated evaluation index data satisfies the predetermined condition (step S50: Yes), the control parameter generation system 1 ends the first control parameter adjustment process.
<考察>
上述したように、上記構成の制御パラメータ生成システム1は、第1の制御パラメータ調整処理を実行することにより、評価指標値データの生成が所定の条件を満たすまで、繰り返し制御パラメータの生成を行うことで制御パラメータの調整を行う。 <Consideration>
As described above, the controlparameter generation system 1 configured as described above repeatedly generates control parameters by executing the first control parameter adjustment process until the generation of evaluation index value data satisfies a predetermined condition. Adjust the control parameters.
上述したように、上記構成の制御パラメータ生成システム1は、第1の制御パラメータ調整処理を実行することにより、評価指標値データの生成が所定の条件を満たすまで、繰り返し制御パラメータの生成を行うことで制御パラメータの調整を行う。 <Consideration>
As described above, the control
このため、制御パラメータ生成システム1によると、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
Therefore, according to the control parameter generation system 1, appropriate control parameters can be generated in a relatively short time without the need for skilled engineers.
(実施の形態2)
以下、実施の形態1に係る制御パラメータ生成システム1から、一部の構成が変更されて構成される実施の形態2に係る制御パラメータ生成システムについて説明する。 (Embodiment 2)
Hereinafter, a control parameter generation system according to a second embodiment, which is configured by partially changing the configuration of the controlparameter generation system 1 according to the first embodiment, will be described.
以下、実施の形態1に係る制御パラメータ生成システム1から、一部の構成が変更されて構成される実施の形態2に係る制御パラメータ生成システムについて説明する。 (Embodiment 2)
Hereinafter, a control parameter generation system according to a second embodiment, which is configured by partially changing the configuration of the control
制御パラメータ生成システム1は、センサ30が、駆動対象物24の位置を測定して測定データを出力し、生成装置10が、センサ30から出力された測定データを取得するシステムの一例であった。
The control parameter generation system 1 was an example of a system in which the sensor 30 measures the position of the driven object 24 and outputs measurement data, and the generation device 10 acquires the measurement data output from the sensor 30.
これに対して、実施の形態2に係る制御パラメータ生成システムは、実施の形態2に係る処理装置付きセンサが、駆動対象物24の位置を測定して評価指標データを出力し、実施の形態2に係る生成装置が、実施の形態2に係る処理装置付きセンサから出力された評価指標データを取得するシステムの一例となっている。
On the other hand, in the control parameter generation system according to the second embodiment, the sensor with the processing device according to the second embodiment measures the position of the driven object 24 and outputs evaluation index data. The generation device according to Embodiment 2 is an example of a system that acquires evaluation index data output from the sensor with a processing device according to the second embodiment.
ここでは、実施の形態2に係る制御パラメータ生成システムについて、制御パラメータ生成システム1と同様の構成要素については、既に説明済みであるとして同じ符号を振ってその詳細な説明を省略し、制御パラメータ生成システム1との相違点を中心に説明する。
Here, regarding the control parameter generation system according to Embodiment 2, the same components as those in control parameter generation system 1 are given the same reference numerals as those in control parameter generation system 1, and the detailed explanation thereof will be omitted. The differences from System 1 will be mainly explained.
図11は、実施の形態2に係る制御パラメータ生成システム1Aの概要を示す模式図である。
FIG. 11 is a schematic diagram showing an overview of a control parameter generation system 1A according to the second embodiment.
図12は、制御パラメータ生成システム1Aの構成を示すブロック図である。
FIG. 12 is a block diagram showing the configuration of the control parameter generation system 1A.
図11に示すように、制御パラメータ生成システム1Aは、実施の形態1に係る制御パラメータ生成システム1から、生成装置10が生成装置10Aに変更され、センサ30が処理装置付きセンサ30Aに変更されて構成される。また、図12に示すように、生成装置10Aは、生成装置10から、入力部11が入力部11Aに変更され、制御回路13が制御回路13Aに変更されて構成される。
As shown in FIG. 11, the control parameter generation system 1A differs from the control parameter generation system 1 according to the first embodiment in that the generation device 10 is changed to a generation device 10A, and the sensor 30 is changed to a sensor with a processing device 30A. configured. Further, as shown in FIG. 12, the generation device 10A is configured by changing the input unit 11 from the generation device 10 to an input unit 11A, and changing the control circuit 13 to a control circuit 13A.
処理装置付きセンサ30Aは、駆動対象物24の位置を測定する。そして、処理装置付きセンサ30Aは、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを生成する。さらに、処理装置付きセンサ30Aは、生成した測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを生成し、生成した評価指標データを、生成装置10Aに出力する。
The sensor 30A with a processing device measures the position of the driven object 24. Then, the processing device-equipped sensor 30A generates measurement data representing the position of the driven object 24 after the time when the driven object 24 reaches a permissible position where it can be evaluated that the driven object 24 has reached a predetermined target position. Furthermore, the sensor with processing device 30A generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the generated measurement data, and outputs the generated evaluation index data to the generation device 10A.
処理装置付きセンサ30Aは、例えば、駆動対象物24の位置を測定するセンシングデバイスと、プロセッサと、メモリと、入出力インターフェースとを備えるコンピュータ装置において、プロセッサがメモリに記憶されるプログラムを実行することで実現される。このようなコンピュータ装置は、例えばパーソナルコンピュータ(パソコン)である。
The sensor with processing device 30A is, for example, a computer device including a sensing device that measures the position of the driven object 24, a processor, a memory, and an input/output interface, in which the processor executes a program stored in the memory. It is realized by Such a computer device is, for example, a personal computer (personal computer).
または、処理装置付きセンサ30Aは、例えば、画像処理装置を含むとしてもよい。この場合、処理装置付きセンサ30Aは、例えば、画像処理装置を用いて駆動対象物24の画像に対して画像処理を行うことで、駆動対象物24の位置を測定するとしてもよい。
Alternatively, the processing device-equipped sensor 30A may include, for example, an image processing device. In this case, the sensor with processing device 30A may measure the position of the driven object 24 by performing image processing on the image of the driven object 24 using an image processing device, for example.
入力部11Aは、処理装置付きセンサ30Aから出力された評価指標データを取得する。
The input unit 11A acquires the evaluation index data output from the sensor with processing device 30A.
制御回路13Aは、入力部11Aにより取得された評価指標データに基づいて、評価指標データと制御パラメータとの関係を学習する機械学習モデル14を用いて、制御パラメータを更新する。
The control circuit 13A updates the control parameters using the machine learning model 14 that learns the relationship between the evaluation index data and the control parameters based on the evaluation index data acquired by the input unit 11A.
<動作>
以下、上記構成の制御パラメータ生成システム1Aが行う動作について説明する。 <Operation>
The operations performed by the control parameter generation system 1A having the above configuration will be described below.
以下、上記構成の制御パラメータ生成システム1Aが行う動作について説明する。 <Operation>
The operations performed by the control parameter generation system 1A having the above configuration will be described below.
制御パラメータ生成システム1Aは、実施の形態1に係る第1の制御パラメータ調整処理から、一部の処理が変更された第2の制御パラメータ調整処理を行う。
The control parameter generation system 1A performs a second control parameter adjustment process in which a part of the process is changed from the first control parameter adjustment process according to the first embodiment.
図13は、第2の制御パラメータ調整処理のシーケンス図であり、図14は、第2の制御パラメータ調整処理のフローチャートである。
FIG. 13 is a sequence diagram of the second control parameter adjustment process, and FIG. 14 is a flowchart of the second control parameter adjustment process.
図13、図14に示すように、第2の制御パラメータ調整処理は、ステップS110の処理~ステップS165の処理を含む。これらの処理のうち、ステップS110の処理~ステップS130の処理、および、ステップS150の処理~ステップS165の処理は、それぞれ、第1の制御パラメータ調整処理における、ステップS10の処理~ステップS30の処理、および、ステップS50の処理~ステップS65の処理と同様の処理である。
As shown in FIGS. 13 and 14, the second control parameter adjustment process includes the process from step S110 to step S165. Of these processes, the process from step S110 to step S130 and the process from step S150 to step S165 are the process from step S10 to step S30 in the first control parameter adjustment process, respectively. The process is similar to the process from step S50 to step S65.
このため、ここでは、ステップS135の処理~ステップS145の処理を中心に説明する。
Therefore, here, the processing from step S135 to step S145 will be mainly explained.
ステップS130の処理が終了すると、処理装置付きセンサ30Aは、駆動対象物24の位置を測定し(ステップS135)、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを生成する。そして、処理装置付きセンサ30Aは、生成した測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを生成し(ステップS140)、生成した評価指標データを、入力部11Aに出力する。すると、入力部11Aは、処理装置付きセンサ30Aから出力された評価指標データを取得する(ステップS145)。
When the process in step S130 is completed, the sensor with processing device 30A measures the position of the driven object 24 (step S135), and the time when the driven object 24 reaches an allowable position at which it can be evaluated that it has reached a predetermined target position. Measurement data representing the subsequent position of the driven object 24 is generated. Then, the sensor with processing device 30A generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the generated measurement data (step S140), and inputs the generated evaluation index data to the input section 11A. Output to. Then, the input unit 11A obtains the evaluation index data output from the processing device-equipped sensor 30A (step S145).
ステップS145の処理が終了すると、制御パラメータ生成システム1Aは、ステップS150の処理に進む。
Upon completion of the process in step S145, the control parameter generation system 1A proceeds to the process in step S150.
ステップS150の処理において、評価指標データの生成が、所定の条件を満たしている場合に(ステップS150:Yes)、制御パラメータ生成システム1Aは、その第2の制御パラメータ調整処理を終了する。
In the process of step S150, if the generation of evaluation index data satisfies the predetermined conditions (step S150: Yes), the control parameter generation system 1A ends the second control parameter adjustment process.
<考察>
上述したように、上記構成の制御パラメータ生成システム1Aは、実施の形態1に係る制御パラメータ生成システム1と同様に、第2の制御パラメータ調整処理を実行することにより、評価指標値データの生成が所定の条件を満たすまで、繰り返し制御パラメータの生成を行うことで制御パラメータの調整を行う。 <Consideration>
As described above, similarly to the controlparameter generation system 1 according to the first embodiment, the control parameter generation system 1A having the above configuration can generate evaluation index value data by executing the second control parameter adjustment process. Control parameters are adjusted by repeatedly generating control parameters until predetermined conditions are met.
上述したように、上記構成の制御パラメータ生成システム1Aは、実施の形態1に係る制御パラメータ生成システム1と同様に、第2の制御パラメータ調整処理を実行することにより、評価指標値データの生成が所定の条件を満たすまで、繰り返し制御パラメータの生成を行うことで制御パラメータの調整を行う。 <Consideration>
As described above, similarly to the control
このため、制御パラメータ生成システム1Aによると、制御パラメータ生成システム1と同様に、熟練技術者によることなく、また、比較的短時間で、適正な制御パラメータを生成することができる。
Therefore, like the control parameter generation system 1, the control parameter generation system 1A can generate appropriate control parameters in a relatively short period of time without requiring a skilled engineer.
(実施の形態3)
以下、実施の形態1に係る制御パラメータ生成システム1から、一部の構成が変更されて構成される実施の形態3に係る制御パラメータ生成システム1Bについて説明する。 (Embodiment 3)
Hereinafter, a controlparameter generation system 1B according to a third embodiment, which is configured by partially changing the configuration of the control parameter generation system 1 according to the first embodiment, will be described.
以下、実施の形態1に係る制御パラメータ生成システム1から、一部の構成が変更されて構成される実施の形態3に係る制御パラメータ生成システム1Bについて説明する。 (Embodiment 3)
Hereinafter, a control
実施の形態3に係る制御パラメータ生成システムは、以下に示す制御パラメータの再調整を行うことができるシステムの一例である。すなわち、制御パラメータ生成システムは、評価指標データの生成が所定の条件を満たすまで繰り返し制御パラメータの生成を行うことで制御パラメータの調整を行う。そして、調整後の評価指標データが、制御パラメータ生成システムを利用するユーザの満足のいくものでなかったときには、そのユーザにより指定された時間帯を除外する時間帯に生成された評価指標データに基づいて、制御パラメータの再調整を行う。
The control parameter generation system according to the third embodiment is an example of a system that can readjust the control parameters described below. That is, the control parameter generation system adjusts the control parameters by repeatedly generating the control parameters until the generation of the evaluation index data satisfies a predetermined condition. If the adjusted evaluation index data is not satisfactory to the user using the control parameter generation system, the control parameter generation system is based on evaluation index data generated during a time period excluding the time period specified by the user. Then readjust the control parameters.
ここでは、実施の形態3に係る制御パラメータ生成システム1Bについて、制御パラメータ生成システム1と同様の構成要素については、既に説明済みであるとして同じ符号を振ってその詳細な説明を省略し、制御パラメータ生成システム1との相違点を中心に説明する。
Here, regarding the control parameter generation system 1B according to Embodiment 3, the same components as those in the control parameter generation system 1 will be given the same reference numerals as they have already been explained, and detailed explanation thereof will be omitted. The differences from generation system 1 will be mainly explained.
図15は、実施の形態3に係る制御パラメータ生成システム1Bの概要を示す模式図である。
FIG. 15 is a schematic diagram showing an overview of the control parameter generation system 1B according to the third embodiment.
図16は、制御パラメータ生成システム1Bの構成を示すブロック図である。
FIG. 16 is a block diagram showing the configuration of the control parameter generation system 1B.
図15に示すように、制御パラメータ生成システム1Bは、実施の形態1に係る制御パラメータ生成システム1から、生成装置10が生成装置10Bに変更され、表示装置40が追加されて構成される。表示装置40は、生成装置10Bとは別に備えられてもよいが、生成装置10Bに含まれてよい。また、表示装置40と表示部17とが同一であってもよい。表示装置40が表示部17の一部として含まれる場合として、例えば表示部17が複数の表示装置からなり、表示装置40がそのうちの1つであってもよい。また、表示部17に、表示装置40が示す情報を表示させてもよい。
As shown in FIG. 15, the control parameter generation system 1B is configured by changing the generation device 10 from the control parameter generation system 1 according to Embodiment 1 to a generation device 10B, and adding a display device 40. The display device 40 may be provided separately from the generation device 10B, or may be included in the generation device 10B. Further, the display device 40 and the display section 17 may be the same. In a case where the display device 40 is included as part of the display unit 17, for example, the display unit 17 may include a plurality of display devices, and the display device 40 may be one of them. Further, information shown by the display device 40 may be displayed on the display unit 17.
また、図16に示すように、生成装置10Bは、生成装置10から、出力部12が出力部12Bに変更され、制御回路13が制御回路13Bに変更されて構成される。
Further, as shown in FIG. 16, the generation device 10B is configured by changing the output unit 12 from the generation device 10 to an output unit 12B and changing the control circuit 13 to a control circuit 13B.
制御回路13Bは、制御回路13が行う動作と同様の動作を行い、さらに、以下の動作を行う。
The control circuit 13B performs operations similar to those performed by the control circuit 13, and further performs the following operations.
制御回路13Bは、制御パラメータ生成システム1Bにより、評価指標データの生成が所定の条件を満たすまで繰り返し制御パラメータの生成を行って、制御パラメータの調整が行われる。その後、制御パラメータ生成システム1Bを利用するユーザから、少なくとも一の時間帯が指定されると共に制御パラメータの更新の再開が指示される。当該指示があると、その少なくとも一の時間帯において新たに生成する評価指標データ、および、新たに生成する評価指標データ以外については、既に生成された評価指標データに基づいて、機械学習モデル14を用いて、更新された制御パラメータを再更新する。
The control circuit 13B adjusts the control parameters by repeatedly generating control parameters using the control parameter generation system 1B until the generation of evaluation index data satisfies a predetermined condition. Thereafter, the user using the control parameter generation system 1B specifies at least one time period and instructs to restart updating of the control parameters. When the instruction is given, the machine learning model 14 is executed based on the evaluation index data that is newly generated in at least one time period and the evaluation index data that has already been generated, except for the evaluation index data that is newly generated. to re-update the updated control parameters.
制御回路13Bが行う更新された制御パラメータの再更新の具体的な動作例については、後述する第3の制御パラメータ調整処理を説明する中で説明する。
A specific example of the operation of re-updating the updated control parameters performed by the control circuit 13B will be explained in the description of the third control parameter adjustment process described later.
制御回路13Bは、評価指標データを生成する際に、評価指標データを、その評価指標データの生成時刻と対応付けて生成する。そして、制御回路13Bは、それら生成した評価指標データと、評価指標データの生成時刻とに基づいて、調整履歴を更新する。
When generating evaluation index data, the control circuit 13B generates the evaluation index data in association with the generation time of the evaluation index data. Then, the control circuit 13B updates the adjustment history based on the generated evaluation index data and the generation time of the evaluation index data.
図17は、制御回路13Bが更新する調整履歴のデータ構成の一例を示すデータ構成図である。
FIG. 17 is a data configuration diagram showing an example of the data configuration of the adjustment history updated by the control circuit 13B.
図17に示すように、調整履歴は、一例として、更新回数と、タイムスタンプと、最良評価値と、最良評価値時の制御パラメータ値とが対応付けられて構成される。
As shown in FIG. 17, the adjustment history is configured by associating, for example, the number of updates, a time stamp, the best evaluation value, and the control parameter value at the time of the best evaluation value.
更新回数は、制御パラメータ生成システム1Bにおいて制御パラメータの調整が開始された後において、制御回路13Bが制御パラメータを生成した回数、すなわち、制御回路13Bが制御パラメータを更新した回数である。
The number of updates is the number of times the control circuit 13B has generated the control parameters, that is, the number of times the control circuit 13B has updated the control parameters after the control parameter generation system 1B starts adjusting the control parameters.
タイムスタンプは、制御回路13Bが、対応する更新回数が示す回数の評価指標データを生成した際の、その評価指標データの生成時刻である。
The time stamp is the generation time of the evaluation index data when the control circuit 13B generated the evaluation index data the number of times indicated by the corresponding number of updates.
最良評価値は、制御パラメータ生成システム1Bにおいて制御パラメータの調整が開始された後において、対応するタイムスタンプが示す時刻までに、制御回路13Bが生成した最良の評価指標データの値(ここでは、最小の評価指標データの値)である。
The best evaluation value is the value of the best evaluation index data (here, the minimum (the value of the evaluation index data).
最良評価値時の制御パラメータ値は、制御回路13Bが対応する最良評価値を算出した際の制御パラメータの値である。
The control parameter value at the time of the best evaluation value is the value of the control parameter when the control circuit 13B calculates the corresponding best evaluation value.
再び、図15、図16に戻って、制御パラメータ生成システム1Bの説明を続ける。
Returning again to FIGS. 15 and 16, the description of the control parameter generation system 1B will be continued.
制御回路13Bは、評価指標データを生成すると、生成した評価指標データが示す駆動対象物24の振動が所定の閾値を超えたか否かを判定する。そして、制御回路13Bは、その調べた結果に基づいて、閾値判定履歴を更新する。
After generating the evaluation index data, the control circuit 13B determines whether the vibration of the driven object 24 indicated by the generated evaluation index data exceeds a predetermined threshold. Then, the control circuit 13B updates the threshold determination history based on the results of the investigation.
図18は、制御回路13Bが更新する閾値判定履歴のデータ構成の一例を示すデータ構成図である。
FIG. 18 is a data configuration diagram showing an example of the data configuration of the threshold determination history updated by the control circuit 13B.
図18に示すように、閾値判定履歴は、一例として、更新回数と、タイムスタンプと、閾値に対する比率と、判定結果とが対応付けられて構成される。
As shown in FIG. 18, the threshold determination history is configured by associating, for example, the number of updates, a timestamp, a ratio to the threshold, and a determination result.
更新回数とタイムスタンプとは、図17に示す調整履歴における更新回数とタイムスタンプと同様である。
The number of updates and the time stamp are the same as the number of updates and the time stamp in the adjustment history shown in FIG.
閾値に対する比率は、対応する更新回数が示す回数の評価指標データを生成した際の、その評価指標データが示す駆動対象物24の振動の、所定の閾値に対する比率である。
The ratio to the threshold is the ratio of the vibration of the driven object 24 indicated by the evaluation index data to the predetermined threshold when the evaluation index data is generated the number of times indicated by the corresponding number of updates.
判定結果は、対応する閾値に対する比率が1を超える場合に「True」を示し、1を超えない場合に「False」を示す情報である。判定結果は、例えば、対応する閾値に対する比率が1を超えるときに論理値「1」となり、1を超えないのときに論理値「0」となる1ビットの信号である。
The determination result is information that indicates "True" when the ratio to the corresponding threshold value exceeds 1, and indicates "False" when the ratio does not exceed 1. The determination result is, for example, a 1-bit signal that has a logical value of "1" when the ratio to the corresponding threshold value exceeds 1, and has a logical value of "0" when it does not exceed 1.
再び、図15、図16に戻って、制御パラメータ生成システム1Bの説明を続ける。
Returning again to FIGS. 15 and 16, the description of the control parameter generation system 1B will be continued.
制御回路13Bは、評価指標データの生成が一旦所定の条件を満たした場合、すなわち、制御パラメータの調整が一旦終了した場合、閾値判定履歴を参照して、判定結果が「Faulse」に対応付けられた更新回数回目に更新された評価指標データを優良評価指標データと特定し、判定結果が「True」に対応付けられた更新回数回目に更新された評価指標データを不良評価指標データと特定する。
Once the generation of the evaluation index data satisfies a predetermined condition, that is, once the adjustment of the control parameters has been completed, the control circuit 13B refers to the threshold value determination history and associates the determination result with "Fault". The evaluation index data that was updated for the first update number is specified as good evaluation index data, and the evaluation index data that was updated for the first update number for which the determination result is "True" is specified as poor evaluation index data.
ここで、前述した通り、優良評価指標データは、評価指標データが示す駆動対象物24の振動が所定の閾値を超えていない評価指標データであり、不良評価指標データは、評価指標データが示す駆動対象物24の振動が所定の閾値を超えている評価指標データである。
Here, as described above, the excellent evaluation index data is evaluation index data in which the vibration of the driven object 24 indicated by the evaluation index data does not exceed a predetermined threshold value, and the poor evaluation index data is the This is evaluation index data indicating that the vibration of the object 24 exceeds a predetermined threshold.
そして、制御回路13Bは、優良評価指標データの数の回数だけ、評価指標データの更新を再度行う場合に要する予想時間を算出する。
Then, the control circuit 13B calculates the expected time required to update the evaluation index data again as many times as the number of excellent evaluation index data.
出力部12Bは、出力部12が行う動作と同様の動作を行い、さらに、以下の動作を行う。
The output unit 12B performs the same operations as the output unit 12, and further performs the following operations.
出力部12Bは、評価指標データが示す駆動対象物24の振動が所定の閾値を超えた1以上の時間帯を示す時刻データを、表示装置40に出力する。ここでは、出力部12Bは、調整履歴と閾値判定履歴とを表示装置40に出力することで、時刻データを表示装置40に出力する。この際、出力部12Bは、制御回路13Bにより算出された予想時間を示す付加情報を表示装置40に出力する。
The output unit 12B outputs time data indicating one or more time periods in which the vibration of the driven object 24 indicated by the evaluation index data exceeds a predetermined threshold value to the display device 40. Here, the output unit 12B outputs time data to the display device 40 by outputting the adjustment history and the threshold value determination history to the display device 40. At this time, the output unit 12B outputs additional information indicating the expected time calculated by the control circuit 13B to the display device 40.
表示装置40は、出力部12Bから出力されるデータおよび情報に基づく画像を生成して表示する。そして、表示装置40は、表示する画像に基づいて行われる、制御パラメータ生成システム1Bを利用するユーザによる入力を受け付ける。
The display device 40 generates and displays an image based on the data and information output from the output unit 12B. The display device 40 then receives input from a user using the control parameter generation system 1B based on the displayed image.
図19は、表示装置40が表示する画像の一例である。図19は、制御パラメータ生成システム1Bを利用するユーザに対して、制御パラメータを再調整する場合における、再調整の対象から除外する評価指標データが生成された時間帯の指定を促す画像(以下、「除外対象区間指定画像」とも称する)の一例となっている。
FIG. 19 is an example of an image displayed by the display device 40. FIG. 19 shows an image (hereinafter referred to as This is an example of an image (also referred to as an "exclusion target section designation image").
図19において、横軸は、制御パラメータの更新回数、または、制御パラメータの調整を開始してからの経過時間(調整時間)であり、縦軸は、最良評価値である。
In FIG. 19, the horizontal axis is the number of updates of the control parameters or the elapsed time (adjustment time) since the start of adjustment of the control parameters, and the vertical axis is the best evaluation value.
図19において、除外対象区間は、制御パラメータ生成システム1Bを利用するユーザに対して、機械学習モデルを再学習させる際において、再学習の対象とする評価指標データから除外することを推奨する評価指標データが生成された時間帯である。ここでは、除外対象区間は、1以上の不良評価指標データが連続して生成された時間帯となっている。
In FIG. 19, the exclusion target section is an evaluation index that is recommended to be excluded from evaluation index data to be retrained when relearning a machine learning model for a user using the control parameter generation system 1B. The time period in which the data was generated. Here, the exclusion target section is a time period in which one or more pieces of defective evaluation index data are continuously generated.
図19において、予想調整時間は、制御パラメータを再調整する場合における、予想される再調整時間である。ここでは、予想調整時間は、制御回路13によって算出された予想時間となっている。
In FIG. 19, the expected adjustment time is the expected readjustment time when the control parameters are readjusted. Here, the expected adjustment time is the expected time calculated by the control circuit 13.
また、図19における破線は、制御パラメータの調整が理想的に進行した場合における最良評価値の推移を示す仮想的な線であって、実際には、調整再開時点指定画像には表示されない。
Furthermore, the broken line in FIG. 19 is a virtual line indicating the transition of the best evaluation value when the control parameter adjustment progresses ideally, and is not actually displayed in the adjustment restart point designation image.
ユーザは、例えば、除外対象区間指定画像に対して、「除外対象区間全選択」の横のボタンをクリックすることで、表示する除外対象区間の全ての区間に対応する評価指標データを、機械学習モデルを再学習させる際において、再学習の対象とする評価指標データから除外させることができる。
For example, by clicking the button next to "Select all excluded sections" for the exclusion target section specified image, the user can apply machine learning to the evaluation index data corresponding to all sections of the excluded sections to be displayed. When relearning a model, it can be excluded from evaluation index data targeted for relearning.
また、ユーザは、例えば、除外対象区間指定画像に対して、「除外対象区間選択」の横のボタンをクリックした上で、所望の除外対象区間を選択(例えば、クリック)することで、所望の除外対象区間に対応する評価指標データを、機械学習モデルを再学習させる際において、再学習の対象とする評価指標データから除外させることができる。この場合、制御回路13Bは、所望の除外対象区間として選択されなかった除外対象区間に対応する不良評価指標データを、改めて、優良評価指標データとして特定しなおす。これにより、所望の除外対象区間として選択された除外対象区間に対応する評価指標データのみが不良評価指標データとして特定され、それ以外の評価指標データが優良評価指標データとして特定されることとなる。また、この場合、制御回路13Bは、新たに特定された優良評価指標データを用いて、予想時間を算出しなおすとしてもよい。そして、表示装置40は、算出しなおされた予想時間で、表示する予想調整時間を更新するとしてもよい。
In addition, the user can, for example, click the button next to "Select Exclusion Target Section" on the exclusion target section designation image, and then select (for example, click) the desired exclusion target section to select the desired exclusion target section. The evaluation index data corresponding to the exclusion target section can be excluded from the evaluation index data to be retrained when retraining the machine learning model. In this case, the control circuit 13B re-identifies the poor evaluation index data corresponding to the exclusion target section that was not selected as the desired exclusion target section as good evaluation index data. As a result, only the evaluation index data corresponding to the exclusion target section selected as the desired exclusion target section is specified as bad evaluation index data, and the other evaluation index data is specified as excellent evaluation index data. Further, in this case, the control circuit 13B may recalculate the expected time using the newly specified excellent evaluation index data. The display device 40 may then update the displayed expected adjustment time with the recalculated expected time.
<動作>
以下、上記構成の制御パラメータ生成システム1Bが行う動作について説明する。 <Operation>
The operations performed by the controlparameter generation system 1B having the above configuration will be described below.
以下、上記構成の制御パラメータ生成システム1Bが行う動作について説明する。 <Operation>
The operations performed by the control
制御パラメータ生成システム1Bは、実施の形態1に係る第1の制御パラメータ調整処理から、一部の処理が変更された第3の制御パラメータ調整処理を行う。
The control parameter generation system 1B performs a third control parameter adjustment process in which a part of the process is changed from the first control parameter adjustment process according to the first embodiment.
図20、図21は、第3の制御パラメータ調整処理のフローチャートである。
20 and 21 are flowcharts of the third control parameter adjustment process.
図20、図21に示すように、第3の制御パラメータ調整処理は、ステップS210の処理~ステップS297の処理を含む。これらの処理のうち、ステップS210の処理~ステップS240の処理、および、ステップS250の処理~ステップS265の処理は、それぞれ、第1の制御パラメータ調整処理における、ステップS10の処理~ステップS40の処理、および、ステップS50の処理~ステップS65の処理と同様の処理である。
As shown in FIGS. 20 and 21, the third control parameter adjustment process includes the process from step S210 to step S297. Among these processes, the process from step S210 to step S240 and the process from step S250 to step S265 are the process from step S10 to step S40 in the first control parameter adjustment process, respectively. The process is similar to the process from step S50 to step S65.
このため、ここでは、ステップS245の処理~ステップS247の処理、および、ステップS270の処理~ステップS297の処理を中心に説明する。
Therefore, here, the processing from step S245 to step S247 and the processing from step S270 to step S297 will be mainly described.
ステップS240の処理が終了すると、制御回路13Bは、測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを、評価指標データの生成時刻と対応付けて生成する(ステップS245)。そして、制御回路13Bは、それら生成した評価指標データと、評価指標データの生成時刻とに基づいて、調整履歴を更新する(ステップS246)。
When the process of step S240 is completed, the control circuit 13B generates evaluation index data indicating the vibration of the driven object 24 after the settling time based on the measurement data in association with the generation time of the evaluation index data (step S245). Then, the control circuit 13B updates the adjustment history based on the generated evaluation index data and the generation time of the evaluation index data (step S246).
評価指標データを生成すると、制御回路13Bは、生成した評価指標データが示す駆動対象物24の振動が所定の閾値を超えたか否かを判定する。そして、制御回路13Bは、その調べた結果に基づいて、閾値判定履歴を更新する(ステップS247)。
After generating the evaluation index data, the control circuit 13B determines whether the vibration of the driven object 24 indicated by the generated evaluation index data exceeds a predetermined threshold. Then, the control circuit 13B updates the threshold determination history based on the results of the investigation (step S247).
ステップS247の処理が終了すると、制御パラメータ生成システム1Bは、ステップS250の処理に進む。
Upon completion of the process in step S247, the control parameter generation system 1B proceeds to the process in step S250.
ステップS250の処理において、評価指標データの生成が、所定の条件を満たしている場合に(ステップS250:Yes)、制御回路13Bは、予想時間を算出する(ステップS270)。そして、出力部12Bは、評価指標データが示す駆動対象物24の振動が所定の閾値を超えた1以上の時間帯を示す時刻データを、予想時間を示す付加情報と共に表示装置40に出力する(ステップS271)。すると、表示装置40は、これらデータおよび情報に基づいて除外対象区間指定画像を生成して表示する(ステップS272)。
In the process of step S250, if the generation of evaluation index data satisfies a predetermined condition (step S250: Yes), the control circuit 13B calculates the expected time (step S270). Then, the output unit 12B outputs time data indicating one or more time periods in which the vibration of the driven object 24 indicated by the evaluation index data exceeds a predetermined threshold value to the display device 40 together with additional information indicating the expected time ( Step S271). Then, the display device 40 generates and displays an exclusion target section designation image based on these data and information (step S272).
除外対象区間指定画像を表示すると、表示装置40は、表示する除外対象区間指定画像に基づいて行われる、制御パラメータ生成システム1Bを利用するユーザによる入力を受け付ける(ステップS275)。
After displaying the exclusion target section designation image, the display device 40 accepts input by the user using the control parameter generation system 1B based on the displayed exclusion target section designation image (step S275).
表示装置40によって受け付けられた入力が、制御パラメータを再調整する場合における除外対象区間を指定する旨の入力であった場合(ステップS280:Yes)、制御回路13Bは、機械学習モデル14の学習の度合いを初期化した上で、すなわち、第3の制御パラメータ調整処理の開始時点に巻き戻した上で、機械学習モデル14に、優良評価指標データで、評価指標データと、対応制御パラメータとの関係を再学習させる(ステップS285)。
If the input received by the display device 40 is an input specifying an exclusion target section when readjusting the control parameters (step S280: Yes), the control circuit 13B controls the learning of the machine learning model 14. After initializing the degree, that is, rewinding to the start point of the third control parameter adjustment process, the machine learning model 14 is provided with excellent evaluation index data and the relationship between the evaluation index data and the corresponding control parameter. is re-learned (step S285).
機械学習モデル14の再学習が終了すると、制御回路13Bは、未選択の優良評価指標データが存在するか否かを調べる(ステップS286)。ここで、未選択の優良評価指標データとは、ステップS286:Yesの処理から、後述するステップS297の処理を経由して再びステップS286の処理へと進むループ処理において、未だ選択されたことのない優良評価指標データである。
When the relearning of the machine learning model 14 is completed, the control circuit 13B checks whether unselected excellent evaluation index data exists (step S286). Here, unselected excellent evaluation index data refers to data that has not yet been selected in a loop process that proceeds from the process of step S286: Yes to the process of step S297, which will be described later, and then returns to the process of step S286. This is excellent evaluation index data.
ステップS286の処理において、未選択の優良評価指標データが存在する場合に(ステップS286:Yes)、制御回路13Bは、未選択の優良評価指標データのうちの1の優良評価指標データを選択する(ステップS287)。そして、制御回路13Bは、選択した優良評価指標データに対応する制御パラメータで、制御パラメータを更新する(ステップS288)。すると、出力部12Bは、制御回路13Bにより新たに更新された制御パラメータを、メモリ21に送信する(ステップS289)。
In the process of step S286, if there is unselected excellent evaluation index data (step S286: Yes), the control circuit 13B selects one of the unselected excellent evaluation index data ( Step S287). Then, the control circuit 13B updates the control parameters with the control parameters corresponding to the selected excellent evaluation index data (step S288). Then, the output unit 12B transmits the control parameters newly updated by the control circuit 13B to the memory 21 (step S289).
ステップS289の処理が終了すると、制御パラメータ生成システム1Bは、ステップS290の処理に進む。
Upon completion of the process in step S289, the control parameter generation system 1B proceeds to the process in step S290.
ステップS290の処理~ステップS296の処理は、それぞれ、ステップS220の処理~ステップS246の処理と同様の処理である。
The processing from step S290 to step S296 is similar to the processing from step S220 to step S246, respectively.
ステップS296の処理が終了すると、機械学習モデル14は、生成した評価指標データと、対応する制御パラメータとの関係を学習する(ステップS297)。
When the process of step S296 is completed, the machine learning model 14 learns the relationship between the generated evaluation index data and the corresponding control parameter (step S297).
ステップS297の処理が終了すると、制御パラメータ生成システム1Bは、ステップS286の処理に進む。
Upon completion of the process in step S297, the control parameter generation system 1B proceeds to the process in step S286.
ステップS280の処理において、表示装置40によって受け付けられた入力が、制御パラメータを再調整する場合における除外対象区間を指定する旨の入力でなかった場合(ステップS280:No)と、ステップS286の処理において、未選択の優良評価指標データが存在しない場合(ステップS286:No)とに、制御パラメータ生成システム1Bは、その第3の制御パラメータ調整処理を終了する。
In the process of step S280, if the input received by the display device 40 is not an input to specify an exclusion target section when readjusting the control parameters (step S280: No), in the process of step S286 If there is no unselected excellent evaluation index data (step S286: No), the control parameter generation system 1B ends the third control parameter adjustment process.
<考察>
上述したように、上記構成の制御パラメータ生成システム1Bは、制御パラメータの調整をやり直す必要がある場合でも、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて制御パラメータを再更新する。このようにすることで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。 <Consideration>
As described above, the controlparameter generation system 1B having the above configuration is configured such that even if it is necessary to re-adjust the control parameters, the control parameter generation system 1B generates a new one in at least one time period in which the vibration indicated by the evaluation index data exceeds a predetermined threshold. The control parameters are re-updated based on the already generated evaluation index data other than the evaluation index data that has been generated. By doing so, it is possible to shorten the period for re-adjusting the control parameters compared to the case where the control parameters are re-adjusted from the beginning.
上述したように、上記構成の制御パラメータ生成システム1Bは、制御パラメータの調整をやり直す必要がある場合でも、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて制御パラメータを再更新する。このようにすることで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。 <Consideration>
As described above, the control
(実施の形態4)
以下、実施の形態3に係る制御パラメータ生成システム1Bから、一部の構成が変更されて構成される実施の形態4に係る制御パラメータ生成システムについて説明する。 (Embodiment 4)
Hereinafter, a control parameter generation system according to a fourth embodiment, which is configured by partially changing the configuration from the controlparameter generation system 1B according to the third embodiment, will be described.
以下、実施の形態3に係る制御パラメータ生成システム1Bから、一部の構成が変更されて構成される実施の形態4に係る制御パラメータ生成システムについて説明する。 (Embodiment 4)
Hereinafter, a control parameter generation system according to a fourth embodiment, which is configured by partially changing the configuration from the control
制御パラメータ生成システム1Bは、センサ30が、駆動対象物24の位置を測定して測定データを出力し、生成装置10Bが、センサ30から出力された測定データを取得するシステムの一例であった。
The control parameter generation system 1B was an example of a system in which the sensor 30 measures the position of the driven object 24 and outputs measurement data, and the generation device 10B acquires the measurement data output from the sensor 30.
これに対して、実施の形態4に係る制御パラメータ生成システムは、実施の形態4に係る処理装置付きセンサが、駆動対象物24の位置を測定して評価指標データを出力し、実施の形態4に係る生成装置が、実施の形態4に係る処理装置付きセンサから出力された評価指標データを取得するシステムの一例となっている。
On the other hand, in the control parameter generation system according to the fourth embodiment, the sensor with the processing device according to the fourth embodiment measures the position of the driven object 24 and outputs evaluation index data. The generation device according to Embodiment 4 is an example of a system that acquires evaluation index data output from the sensor with a processing device according to the fourth embodiment.
ここでは、実施の形態4に係る制御パラメータ生成システムについて、制御パラメータ生成システム1Bと同様の構成要素については、既に説明済みであるとして同じ符号を振ってその詳細な説明を省略し、制御パラメータ生成システム1Bとの相違点を中心に説明する。
Here, regarding the control parameter generation system according to Embodiment 4, the same components as those in the control parameter generation system 1B will be given the same reference numerals as they have already been explained, and detailed explanation thereof will be omitted. The explanation will focus on the differences from system 1B.
図22は、実施の形態4に係る制御パラメータ生成システム1Cの概要を示す模式図である。
FIG. 22 is a schematic diagram showing an overview of the control parameter generation system 1C according to the fourth embodiment.
図23は、制御パラメータ生成システム1Cの構成を示すブロック図である。
FIG. 23 is a block diagram showing the configuration of the control parameter generation system 1C.
図22に示すように、制御パラメータ生成システム1Cは、実施の形態3に係る制御パラメータ生成システム1Bから、生成装置10Bが生成装置10Cに変更され、センサ30が処理装置付きセンサ30Cに変更されて構成される。また、図23に示すように、生成装置10Cは、生成装置10Bから、入力部11が入力部11Cに変更され、制御回路13Bが制御回路13Cに変更されて構成される。
As shown in FIG. 22, the control parameter generation system 1C differs from the control parameter generation system 1B according to the third embodiment in that the generation device 10B is changed to a generation device 10C, and the sensor 30 is changed to a sensor with a processing device 30C. configured. Further, as shown in FIG. 23, the generation device 10C is configured by changing the input section 11 from the generation device 10B to an input section 11C, and changing the control circuit 13B to a control circuit 13C.
処理装置付きセンサ30Cは、処理装置付きセンサ30Aが行う動作と同様の動作を行い、さらに、以下の動作を行う。
The sensor with processing device 30C performs the same operations as the sensor with processing device 30A, and further performs the following operations.
処理装置付きセンサ30Cは、評価指標データを生成する際に、評価指標データを、その評価指標データの生成時刻と対応付けて生成する。そして、処理装置付きセンサ30Cは、生成した評価指標データを生成装置10Cに出力する際に、生成した評価指標データの生成時刻を、評価指標データに対応付けて出力する。
When the sensor with processing device 30C generates evaluation index data, it generates the evaluation index data in association with the generation time of the evaluation index data. Then, when outputting the generated evaluation index data to the generation device 10C, the processing device equipped sensor 30C outputs the generation time of the generated evaluation index data in association with the evaluation index data.
入力部11Cは、入力部11が行う動作と同様の動作を行い、さらに、以下の動作を行う。
The input unit 11C performs the same operations as the input unit 11, and further performs the following operations.
入力部11Cは、処理装置付きセンサ30Cから出力された評価指標データを、評価指標データの生成時刻と対応付けて取得する。
The input unit 11C acquires the evaluation index data output from the sensor with processing device 30C in association with the generation time of the evaluation index data.
制御回路13Cは、制御回路13Bが行う動作と同様の動作を行い、さらに、以下の動作を行う。
The control circuit 13C performs the same operations as the control circuit 13B, and further performs the following operations.
制御回路13Cは、入力部11Cが処理装置付きセンサ30Cから測定データを取得せずに評価指標データを評価指標データの生成時刻と対応付けて取得する場合には、測定データに基づいて評価指標データをその評価指標データ生成時刻と対応付けて生成する代わりに、入力部11Cが取得した評価指標データをその評価指標データの生成時刻と対応付けて取得する。
When the input unit 11C does not acquire the measurement data from the sensor with processing device 30C and acquires the evaluation index data in association with the generation time of the evaluation index data, the control circuit 13C generates the evaluation index data based on the measurement data. Instead of generating the evaluation index data in association with the generation time of the evaluation index data, the evaluation index data acquired by the input unit 11C is acquired in association with the generation time of the evaluation index data.
<動作>
以下、上記構成の制御パラメータ生成システム1Cが行う動作について説明する。 <Operation>
The operation performed by the controlparameter generation system 1C having the above configuration will be described below.
以下、上記構成の制御パラメータ生成システム1Cが行う動作について説明する。 <Operation>
The operation performed by the control
制御パラメータ生成システム1Cは、実施の形態3に係る第3の制御パラメータ調整処理から、一部の処理が変更された第4の制御パラメータ調整処理を行う。
The control parameter generation system 1C performs a fourth control parameter adjustment process in which a part of the process is changed from the third control parameter adjustment process according to the third embodiment.
図24、図25は、第4の制御パラメータ調整処理のフローチャートである。
24 and 25 are flowcharts of the fourth control parameter adjustment process.
図24、図25に示すように、第4の制御パラメータ調整処理は、ステップS310の処理~ステップS397の処理を含む。これらの処理のうち、ステップS310の処理~ステップS330の処理、および、ステップS347の処理~ステップS397の処理は、それぞれ、第3の制御パラメータ調整処理における、ステップS210の処理~ステップS230の処理、および、ステップS247の処理~ステップS297の処理と同様の処理である。
As shown in FIGS. 24 and 25, the fourth control parameter adjustment process includes the process from step S310 to step S397. Of these processes, the process from step S310 to step S330 and the process from step S347 to step S397 are the process from step S210 to step S230 in the third control parameter adjustment process, respectively. The processing is similar to the processing from step S247 to step S297.
このため、ここでは、ステップS335の処理~ステップS346の処理を中心に説明する。
Therefore, here, the processing from step S335 to step S346 will be mainly described.
ステップS330の処理が終了すると、処理装置付きセンサ30Cは、駆動対象物24の位置を測定し(ステップS335)、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを生成する。そして、処理装置付きセンサ30Cは、生成した測定データに基づいて、整定時刻後の駆動対象物24の振動を示す評価指標データを、その評価指標データの生成時刻と対応付けて生成し(ステップS340)、生成した評価指標データを、その評価指標データの生成時刻と対応付けて入力部11Cに出力する。すると、入力部11Cは、処理装置付きセンサ30Cから出力された評価指標データを、その評価指標データの生成時刻と対応付けて、取得する(ステップS345)。そして、制御回路13Cは、それら取得した評価指標データと、評価指標データの生成時刻とに基づいて、調整履歴を更新する(ステップS346)。
When the process in step S330 is completed, the sensor with processing device 30C measures the position of the driven object 24 (step S335), and determines the time when the driven object 24 reaches an allowable position at which it can be evaluated that it has reached a predetermined target position. Measurement data representing the subsequent position of the driven object 24 is generated. Then, based on the generated measurement data, the processing device-equipped sensor 30C generates evaluation index data indicating the vibration of the driven object 24 after the settling time in association with the generation time of the evaluation index data (step S340 ), and outputs the generated evaluation index data to the input unit 11C in association with the generation time of the evaluation index data. Then, the input unit 11C obtains the evaluation index data output from the processing device equipped sensor 30C in association with the generation time of the evaluation index data (step S345). Then, the control circuit 13C updates the adjustment history based on the acquired evaluation index data and the generation time of the evaluation index data (step S346).
ステップS346の処理が終了すると、制御パラメータ生成システム1Cは、ステップS347の処理に進む。
Upon completion of the process in step S346, the control parameter generation system 1C proceeds to the process in step S347.
<考察>
上記構成の制御パラメータ生成システム1Cは、実施の形態3に係る制御パラメータ生成システム1Bと同様に、制御パラメータの調整をやり直す必要がある場合でも、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて制御パラメータを再更新する。このようにすることで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。 <Consideration>
Similarly to the controlparameter generation system 1B according to the third embodiment, the control parameter generation system 1C having the above configuration is configured such that even if it is necessary to re-adjust the control parameters, the vibration indicated by the evaluation index data exceeds a predetermined threshold. Control parameters are re-updated based on already generated evaluation index data other than evaluation index data newly generated in at least one time period. By doing so, it is possible to shorten the period for re-adjusting the control parameters compared to the case where the control parameters are re-adjusted from the beginning.
上記構成の制御パラメータ生成システム1Cは、実施の形態3に係る制御パラメータ生成システム1Bと同様に、制御パラメータの調整をやり直す必要がある場合でも、評価指標データが示す振動が所定の閾値を超えた少なくとも一の時間帯において新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて制御パラメータを再更新する。このようにすることで、制御パラメータの調整を最初からやりなおす場合に比べて、調整をやり直す期間を短縮することができる。 <Consideration>
Similarly to the control
(補足)
以上のように、本出願において開示する技術の例示として、実施の形態1~4に基づいて説明した。しかしながら、本開示は、これら実施の形態1~4に限定されるものではない。本開示の趣旨を逸脱しない限り、当業者が思いつく各種変形を本実施の形態または本変形例に施したものや、異なる実施の形態または変形例における構成要素を組み合わせて構築される形態も、本開示の1つまたは複数の態様の範囲内に含まれてもよい。 (supplement)
As described above, the description has been made based onEmbodiments 1 to 4 as examples of the technology disclosed in this application. However, the present disclosure is not limited to these Embodiments 1 to 4. Unless departing from the spirit of the present disclosure, various modifications that can be thought of by those skilled in the art may be made to this embodiment or this modification, or forms constructed by combining components of different embodiments or modifications are also applicable to this invention. may fall within the scope of one or more aspects of the disclosure.
以上のように、本出願において開示する技術の例示として、実施の形態1~4に基づいて説明した。しかしながら、本開示は、これら実施の形態1~4に限定されるものではない。本開示の趣旨を逸脱しない限り、当業者が思いつく各種変形を本実施の形態または本変形例に施したものや、異なる実施の形態または変形例における構成要素を組み合わせて構築される形態も、本開示の1つまたは複数の態様の範囲内に含まれてもよい。 (supplement)
As described above, the description has been made based on
(1)実施の形態1において、一例として、生成装置10は1台のコンピュータ装置により実現されるとして説明した。しかしながら、生成装置10は、同様の機能を実現することができれば、必ずしも1台のコンピュータ装置により実現される例に限定される必要はない。生成装置10は、例えば、互いに通信可能な複数台のコンピュータ装置により実現されてもよい。
(1) In the first embodiment, as an example, the generation device 10 was described as being realized by one computer device. However, the generation device 10 does not necessarily need to be implemented by one computer device as long as it can implement similar functions. The generation device 10 may be realized by, for example, a plurality of computer devices that can communicate with each other.
図26は、生成装置10が、互いに通信可能な複数台のコンピュータ装置により実現される場合における、生成装置10の構成の一例を示すブロック図である。
FIG. 26 is a block diagram showing an example of the configuration of the generation device 10 in a case where the generation device 10 is realized by a plurality of computer devices that can communicate with each other.
図26に示すように、生成装置10は、(1)入力部11と、出力部12と、操作受け付け部15と、画像生成部16と、表示部17と、通信部18とを備える第1のコンピュータ装置100と、(2)制御回路13と、機械学習モデル14と、通信部19とを備える第2のコンピュータ装置110と、(3)通信部18と通信部19とを通信可能に接続するネットワーク120とによって実現されるとしてもよい。
As shown in FIG. 26, the generation device 10 includes (1) an input section 11, an output section 12, an operation reception section 15, an image generation section 16, a display section 17, and a communication section 18. (2) a second computer device 110 including a control circuit 13, a machine learning model 14, and a communication unit 19; and (3) a communication unit 18 and a communication unit 19 are communicably connected. It may also be realized by a network 120 that supports
この場合、第1のコンピュータ装置100は、例えば、パーソナルコンピュータ(パソコン)であり、第2のコンピュータ装置110は、サーバであってもよい。
In this case, the first computer device 100 may be, for example, a personal computer (personal computer), and the second computer device 110 may be a server.
(2)実施の形態1において、制御パラメータ生成システム1は、(i)センサ30が、駆動対象物24の位置を測定して、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを出力し、(ii)入力部11が、センサ30から出力された測定データを取得する構成であるとして説明した。
(2) In the first embodiment, the control parameter generation system 1 has (i) a tolerance that allows the sensor 30 to measure the position of the driven object 24 and evaluate that the driven object 24 has reached a predetermined target position; The description has been made assuming that the configuration is such that the measurement data representing the position of the driven object 24 after the time when it reaches the position is output, and (ii) the input unit 11 acquires the measurement data output from the sensor 30.
しかしながら、制御パラメータ生成システム1の上記構成は一例であり、制御パラメータ生成システム1は、入力部11が、駆動対象物24が所定の目標位置に到達したと評価できる許容位置に到達した時刻後の駆動対象物24の位置を表す測定データを取得することができれば、必ずしも、上記構成に限定される必要はない。
However, the above-mentioned configuration of the control parameter generation system 1 is an example, and the control parameter generation system 1 is configured such that the input unit 11 determines that the drive object 24 has reached a permissible position after the time when the driven object 24 has reached a permissible position where it can be evaluated that the driven object 24 has reached a predetermined target position. The configuration is not necessarily limited to the above configuration as long as measurement data representing the position of the driven object 24 can be acquired.
図27は、制御パラメータ生成システム1の他の構成の一例である制御パラメータ生成システム1Dであって、センサ30を備えない制御パラメータ生成システム1Dの構成を示すブロック図である。
FIG. 27 is a block diagram showing the configuration of a control parameter generation system 1D that is an example of another configuration of the control parameter generation system 1 and does not include the sensor 30.
制御パラメータ生成システム1Dにおいて、サーボモータ23は、モータ位置(例えば、サーボモータ23が回転式モータの場合であれば、サーボモータ23の回転位置)を検出するセンサを内蔵している。そして、サーボモータ23は、内蔵するセンサにより検出されたサーボモータ23のモータ位置を表すエンコード値を出力する。すなわち、生産装置20Dはエンコード値を出力する。
In the control parameter generation system 1D, the servo motor 23 has a built-in sensor that detects the motor position (for example, if the servo motor 23 is a rotary motor, the rotational position of the servo motor 23). Then, the servo motor 23 outputs an encoded value representing the motor position of the servo motor 23 detected by the built-in sensor. That is, the production device 20D outputs an encoded value.
なお、一般に、このような、モータの位置を検出するセンサはエンコーダとも称される。また、通常、サーボモータはエンコーダを内蔵している。
Note that such a sensor that detects the position of the motor is generally also referred to as an encoder. Additionally, servo motors usually have a built-in encoder.
ここで、このエンコード値が表すサーボモータ23の位置は、駆動対象物24の位置と一対一の対応関係にある。このため、サーボモータ23が出力するエンコード値は、サーボモータ23の位置を表すと共に、駆動対象物24の位置をも表す。したがって、サーボモータ23が内蔵するセンサは、駆動対象物24の位置をも検出し、駆動対象物24の位置を表すエンコード値を出力するともいえる。
Here, the position of the servo motor 23 represented by this encoded value has a one-to-one correspondence with the position of the driven object 24. Therefore, the encoded value output by the servo motor 23 represents the position of the servo motor 23 and also represents the position of the driven object 24. Therefore, it can be said that the sensor built into the servo motor 23 also detects the position of the driven object 24 and outputs an encoded value representing the position of the driven object 24.
そして、制御パラメータ生成システム1Dにおいて、入力部11は、サーボモータ23から出力されるエンコード値を、測定データとして取得する。
In the control parameter generation system 1D, the input unit 11 acquires the encoded value output from the servo motor 23 as measurement data.
(3)本開示の包括的または具体的な態様は、システム、装置、方法、集積回路、プログラムまたはコンピュータ読み取り可能なCD-ROMなどの非一時的な記録媒体で実現されてもよい。また、システム、装置、方法、集積回路、プログラムおよび非一時的な記録媒体の任意な組み合わせで実現されてもよい。例えば、本開示は、生成装置が行う処理をコンピュータ装置に実行させるためのプログラムとして実現されてもよい。
(3) The general or specific aspects of the present disclosure may be realized in a system, device, method, integrated circuit, program, or non-transitory recording medium such as a computer-readable CD-ROM. Further, the present invention may be realized by any combination of systems, devices, methods, integrated circuits, programs, and non-transitory recording media. For example, the present disclosure may be implemented as a program for causing a computer device to execute the processing performed by the generation device.
(4)本開示の生産装置20、20Dは、生成装置10、10A~10Cを内蔵してもよい。本開示の生産装置20、20Dは、制御パラメータ生成システム1、1A~1Dを内蔵してもよい。また、表示装置40は、生成装置10、10A~10Cから分離されて配置されてもよく、生成装置10、10A~10Cとは独立に配置されてもよい。また、表示装置40は、生産装置20、20Dに組み込まれていてもよい。また、センサ30や処理装置付きセンサ30A、30Cは、生産装置20、20Dから分離されて配置されてもよく、生産装置20、20Dとは独立に配置してもよい。また、センサ30や処理装置付きセンサ30A、30Cは生成装置10、10A~10Cに組み込まれて配置されてもよい。また、本開示の生産装置20、20Dは、製品を生産する装置に限定されず、製品を生産する設備やプラント、工場を含み得る。また、本開示は、生産装置20、20D以外に、搬送装置に対しても適用し得る。また、本開示の制御パラメータ生成システム1、1A~1Dは、製品を生産する装置、設備やプラント、倉庫または工場に対しても適用し得る。
(4) The production devices 20 and 20D of the present disclosure may include the generation devices 10 and 10A to 10C. The production apparatuses 20 and 20D of the present disclosure may include the control parameter generation systems 1 and 1A to 1D. Further, the display device 40 may be placed separately from the generating devices 10, 10A to 10C, or may be placed independently of the generating devices 10, 10A to 10C. Further, the display device 40 may be incorporated into the production equipment 20, 20D. Moreover, the sensor 30 and the sensors 30A and 30C with processing devices may be arranged separately from the production apparatuses 20 and 20D, or may be arranged independently of the production apparatuses 20 and 20D. Further, the sensor 30 and the sensors with processing devices 30A and 30C may be installed and arranged in the generation devices 10 and 10A to 10C. Furthermore, the production devices 20 and 20D of the present disclosure are not limited to devices that produce products, but may include equipment, plants, and factories that produce products. Further, the present disclosure can be applied to a conveyance device in addition to the production devices 20 and 20D. Furthermore, the control parameter generation systems 1, 1A to 1D of the present disclosure can also be applied to equipment, equipment, plants, warehouses, or factories that produce products.
本開示は、制御パラメータを生成するシステム等に広く利用可能である。
The present disclosure can be widely used in systems that generate control parameters, etc.
1、1A、1B、1C、1D 制御パラメータ生成システム
10、10A、10B、10C 生成装置
11、11A、11C 入力部
12、12B 出力部
13、13A、13B、13C 制御回路
14 機械学習モデル
15 操作受け付け部
16 画像生成部
17 表示部
18、19 通信部
20、20D 生産装置
21 メモリ
22 制御回路
23 サーボモータ
24 駆動対象物
30 センサ
30A、30C 処理装置付きセンサ
40 表示装置
100 第1のコンピュータ装置
110 第2のコンピュータ装置
120 ネットワーク 1, 1A, 1B, 1C, 1D Control parameter generation system 10, 10A, 10B, 10C Generation device 11, 11A, 11C Input section 12, 12B Output section 13, 13A, 13B, 13C Control circuit 14 Machine learning model 15 Operation reception Section 16 Image generation section 17 Display section 18, 19 Communication section 20, 20D Production device 21 Memory 22 Control circuit 23 Servo motor 24 Drive target 30 Sensor 30A, 30C Sensor with processing device 40 Display device 100 First computer device 110 No. 2 computer equipment 120 network
10、10A、10B、10C 生成装置
11、11A、11C 入力部
12、12B 出力部
13、13A、13B、13C 制御回路
14 機械学習モデル
15 操作受け付け部
16 画像生成部
17 表示部
18、19 通信部
20、20D 生産装置
21 メモリ
22 制御回路
23 サーボモータ
24 駆動対象物
30 センサ
30A、30C 処理装置付きセンサ
40 表示装置
100 第1のコンピュータ装置
110 第2のコンピュータ装置
120 ネットワーク 1, 1A, 1B, 1C, 1D Control
Claims (14)
- 制御パラメータの生成装置が生成し、生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する方法であって、
前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる前記制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備え、
前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力し、
前記センサから出力される前記位置を表す測定データを取得し、
前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成し、
前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新し、
更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力し、
前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力し、
前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する、
制御パラメータの生成方法。 A method for generating control parameters that is generated by a control parameter generation device, used by a production device, a sensor confirms a state of the production device, and a display device displays at least a part of the control parameter, the method comprising:
The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores the control parameters used by the control circuit to control the servo motor, and a drive target driven by the servo motor. equipped with things,
The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. measuring the position of an object and outputting measurement data representing the position;
obtaining measurement data representing the position output from the sensor;
generating evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position based on the measurement data, in association with a generation time of the evaluation index data;
updating the control parameters using a machine learning model that learns the relationship between the evaluation index data and the control parameters based on the evaluation index data;
outputting the updated control parameters to the production device for storage in the memory;
outputting time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device;
When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. re-updating the updated control parameters using the machine learning model based on already generated evaluation index data;
How to generate control parameters. - 前記機械学習モデルは、ベイズ最適化アルゴリズムを利用して、前記評価指標データと前記制御パラメータとの関係を学習する、
請求項1に記載の制御パラメータの生成方法。 The machine learning model uses a Bayesian optimization algorithm to learn the relationship between the evaluation index data and the control parameter.
The method for generating control parameters according to claim 1. - 前記生産装置は、実装装置、加工装置、または、工作装置のいずれか1つを含む、
請求項1に記載の制御パラメータの生成方法。 The production device includes any one of a mounting device, a processing device, or a machining device.
The method for generating control parameters according to claim 1. - 請求項1から請求項3のいずれか1項に記載の制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラム。 A program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method according to any one of claims 1 to 3.
- 請求項1から請求項3のいずれか1項に記載の制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体。 A recording medium recording a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method according to any one of claims 1 to 3.
- 生産装置が使用してセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置であって、
前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備え、
前記センサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記位置を表す測定データを出力し、
前記制御パラメータの生成装置は、前記センサから出力される前記位置を表す測定データを取得する入力部と、
前記測定データに基づいて、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて生成し、前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する第2の制御回路と、
更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する出力部と、を備え、
前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力し、
前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する、
制御パラメータの生成装置。 A control parameter generation device that is used by a production device to generate control parameters whose sensors check the state of the production device and whose display device displays at least a portion of the control parameters, the control parameter generating device comprising:
The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven;
The sensor detects the driven object after a time when the driven object reaches a permissible position where it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. measuring the position of an object and outputting measurement data representing the position;
The control parameter generation device includes an input unit that acquires measurement data representing the position output from the sensor;
Based on the measurement data, evaluation index data representing the vibration of the driven object after the time when the driven object reaches the permissible position is generated in association with the generation time of the evaluation index data, and the evaluation is performed. a second control circuit that updates the control parameter using a machine learning model that learns the relationship between the evaluation index data and the control parameter based on the index data;
an output unit that outputs the updated control parameters to the production device in order to store them in the memory,
The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device,
When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Re-updating the updated control parameters using the machine learning model based on already generated evaluation index data other than the newly generated evaluation index data;
Control parameter generation device. - 制御パラメータの生成装置が生成し、生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する方法であって、
前記生産装置は、サーボモータと、前記サーボモータを制御する制御回路と、前記制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物とを備え、
前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力し、
前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得し、
前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新し、
更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力し、
前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力し、
前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する、
制御パラメータの生成方法。 A method for generating control parameters that is generated by a control parameter generation device, used by a production device, a sensor with a processing device confirms the state of the production device, and a display device displays at least a part of the control parameter, the method comprising:
The production device includes a servo motor, a control circuit that controls the servo motor, a memory that stores control parameters used by the control circuit to control the servo motor, and a driven object driven by the servo motor. and
The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. measuring the position of the driven object, outputting evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position;
obtaining evaluation index data representing vibrations of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, in association with a generation time of the evaluation index data;
updating the control parameters using a machine learning model that learns the relationship between the evaluation index data and the control parameters based on the evaluation index data;
outputting the updated control parameters to the production device for storage in the memory;
outputting time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device;
When re-updating the updated control parameters, newly generated evaluation index data in at least one of the one or more time periods, and evaluation index data other than the newly generated evaluation index data. re-updating the updated control parameters using the machine learning model based on already generated evaluation index data;
How to generate control parameters. - 前記機械学習モデルは、ベイズ最適化アルゴリズムを利用して、前記評価指標データと前記制御パラメータとの関係を学習する、
請求項7に記載の制御パラメータの生成方法。 The machine learning model uses a Bayesian optimization algorithm to learn the relationship between the evaluation index data and the control parameter.
The method for generating control parameters according to claim 7. - 前記評価指標データは、前記処理装置付きセンサにおいて、前記処理装置付きセンサにて測定された前記駆動対象物の位置に基づいて生成される、
請求項7に記載の制御パラメータの生成方法。 The evaluation index data is generated in the sensor with a processing device based on the position of the driven object measured by the sensor with a processing device.
The method for generating control parameters according to claim 7. - 前記処理装置付きセンサは、画像処理装置を含む、
請求項7に記載の制御パラメータの生成方法。 The sensor with a processing device includes an image processing device.
The method for generating control parameters according to claim 7. - 前記生産装置は、実装装置、加工装置、または、工作装置のいずれか1つを含む、
請求項7に記載の制御パラメータの生成方法。 The production device includes any one of a mounting device, a processing device, or a machining device.
The method for generating control parameters according to claim 7. - 請求項7から請求項11のいずれか1項に記載の制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラム。 A program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method according to any one of claims 7 to 11.
- 請求項7から請求項11のいずれか1項に記載の制御パラメータの生成方法を、前記生産装置に接続された情報処理装置のコンピュータに実行させるためのプログラムを記録した記録媒体。 A recording medium recording a program for causing a computer of an information processing device connected to the production device to execute the control parameter generation method according to any one of claims 7 to 11.
- 生産装置が使用して処理装置付きセンサが前記生産装置の状態を確認し、表示装置が少なくともその一部を表示する制御パラメータを生成する、制御パラメータの生成装置であって、
前記生産装置は、サーボモータと、前記サーボモータを制御する第1の制御回路と、前記第1の制御回路が前記サーボモータを制御する際に用いる制御パラメータを記憶するメモリと、前記サーボモータにより駆動される駆動対象物と、を備え、
前記処理装置付きセンサは、前記駆動対象物の位置を所定の目標位置にするための指令に基づいて前記駆動対象物が前記所定の目標位置に到達したと評価できる許容位置に到達した時刻後の前記駆動対象物の位置を測定し、前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを出力し、
前記制御パラメータの生成装置は、
前記処理装置付きセンサから出力される前記駆動対象物が前記許容位置に到達した時刻後の前記駆動対象物の振動を表す評価指標データを、前記評価指標データの生成時刻と対応付けて取得する入力部と、
前記評価指標データに基づいて、前記評価指標データと前記制御パラメータとの関係を学習する機械学習モデルを用いて、前記制御パラメータを更新する第2の制御回路と、
更新された前記制御パラメータを前記メモリに記憶させるために前記生産装置に出力する出力部と、を備え、
前記出力部は、さらに、前記評価指標データが示す振動が所定の閾値を超えた1以上の時間帯を示す時刻データを前記表示装置に出力し、
前記第2の制御回路は、さらに、前記更新された制御パラメータを再更新する場合、前記1以上の時間帯の中のいずれか少なくとも一の時間帯において新たに生成された評価指標データ、および、前記新たに生成された評価指標データ以外については既に生成された評価指標データに基づいて、前記機械学習モデルを用いて、前記更新された制御パラメータを再更新する、
制御パラメータの生成装置。 A control parameter generation device that is used by a production device to generate control parameters that are used by a sensor with a processing device to check the state of the production device and that are displayed at least in part by a display device,
The production device includes a servo motor, a first control circuit that controls the servo motor, a memory that stores control parameters used by the first control circuit to control the servo motor, and a servo motor that controls the servo motor. A driven object to be driven;
The sensor equipped with a processing device is configured to detect a time after a time when the driven object reaches an allowable position at which it can be evaluated that the driven object has reached the predetermined target position based on a command for changing the position of the driven object to a predetermined target position. measuring the position of the driven object, outputting evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position;
The control parameter generation device includes:
An input that obtains evaluation index data representing vibration of the driven object after a time when the driven object reaches the permissible position, which is output from the sensor with the processing device, in association with a generation time of the evaluation index data. Department and
a second control circuit that updates the control parameter using a machine learning model that learns a relationship between the evaluation index data and the control parameter based on the evaluation index data;
an output unit that outputs the updated control parameters to the production device in order to store them in the memory,
The output unit further outputs time data indicating one or more time periods in which the vibration indicated by the evaluation index data exceeds a predetermined threshold value to the display device,
When re-updating the updated control parameters, the second control circuit further includes evaluation index data newly generated in at least one of the one or more time periods, and Re-updating the updated control parameters using the machine learning model based on already generated evaluation index data other than the newly generated evaluation index data;
Control parameter generation device.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2022-107099 | 2022-07-01 | ||
JP2022107099 | 2022-07-01 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024005137A1 true WO2024005137A1 (en) | 2024-01-04 |
Family
ID=89382494
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2023/024162 WO2024005137A1 (en) | 2022-07-01 | 2023-06-29 | Control parameter generation method, program, recording medium, and control parameter-generating device |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2024005137A1 (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050171705A1 (en) * | 2003-11-18 | 2005-08-04 | Peter Renner | Condition monitoring in technical processes |
JP2020148624A (en) * | 2019-03-13 | 2020-09-17 | 沖電気工業株式会社 | Signal processing device, program and method |
WO2021053784A1 (en) * | 2019-09-19 | 2021-03-25 | 三菱電機株式会社 | Motor control device and motor control method |
-
2023
- 2023-06-29 WO PCT/JP2023/024162 patent/WO2024005137A1/en unknown
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050171705A1 (en) * | 2003-11-18 | 2005-08-04 | Peter Renner | Condition monitoring in technical processes |
JP2020148624A (en) * | 2019-03-13 | 2020-09-17 | 沖電気工業株式会社 | Signal processing device, program and method |
WO2021053784A1 (en) * | 2019-09-19 | 2021-03-25 | 三菱電機株式会社 | Motor control device and motor control method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5890477B2 (en) | Robot program correction system | |
US10668619B2 (en) | Controller and machine learning device | |
JP4952025B2 (en) | Operation control method, operation control apparatus, and operation control system | |
US9887661B2 (en) | Machine learning method and machine learning apparatus learning operating command to electric motor and controller and electric motor apparatus including machine learning apparatus | |
US11592800B2 (en) | Abnormality detector of a manufacturing machine using machine learning | |
US10618164B2 (en) | Robot system having learning control function and learning control method | |
JP6348137B2 (en) | Machining machine system for judging the quality of workpieces | |
US9321177B2 (en) | Reducer abnormality determination method, abnormality determination device, and robot system | |
JP2005310109A (en) | Self-calibrating sensor orientation system | |
US10520912B2 (en) | Robot controller having function that simplifies learning, and robot control method | |
US10338541B2 (en) | Machine learning to establish optimal filter for removing external noise without degrading responsivity | |
JP6928669B2 (en) | Control system, factory system, learning system, estimation model generation method and actuator state estimation method | |
JP6577522B2 (en) | Control device and machine learning device | |
JP2005305633A (en) | Self-calibrating orienting system for operation device | |
US11397414B2 (en) | Mechanical equipment control system, control apparatus for mechanical equipment, and method for controlling mechanical equipment | |
JP6885085B2 (en) | Control device, control method, and control program | |
JP7043801B2 (en) | Abnormality sign notification system, abnormality sign notification method and program | |
US20190311079A1 (en) | Simulation apparatus, robot, simulation method, and program therefor | |
JPWO2009019953A1 (en) | Electric motor control device and gain adjustment method thereof | |
US11077552B2 (en) | Control system, machine learning apparatus, maintenance assistance apparatus, data generating method, and maintenance assisting method | |
JP6946654B2 (en) | Control device, control method, and control program | |
WO2024005137A1 (en) | Control parameter generation method, program, recording medium, and control parameter-generating device | |
JP2012103827A (en) | Control parameter adjustment device, control parameter adjustment method, and program | |
WO2024005136A1 (en) | Method for generating control parameter, program, recording medium, and device for generating control parameter | |
WO2024005138A1 (en) | Control parameter generation method, program, recording medium, and control parameter generation device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23831576 Country of ref document: EP Kind code of ref document: A1 |