WO2021192023A1 - 数値制御装置 - Google Patents

数値制御装置 Download PDF

Info

Publication number
WO2021192023A1
WO2021192023A1 PCT/JP2020/012983 JP2020012983W WO2021192023A1 WO 2021192023 A1 WO2021192023 A1 WO 2021192023A1 JP 2020012983 W JP2020012983 W JP 2020012983W WO 2021192023 A1 WO2021192023 A1 WO 2021192023A1
Authority
WO
WIPO (PCT)
Prior art keywords
position command
auxiliary function
unit
numerical control
vibration
Prior art date
Application number
PCT/JP2020/012983
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
泰一 石田
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to JP2020542180A priority Critical patent/JP6775720B1/ja
Priority to CN202080098600.XA priority patent/CN115315665A/zh
Priority to DE112020006576.6T priority patent/DE112020006576T5/de
Priority to PCT/JP2020/012983 priority patent/WO2021192023A1/ja
Publication of WO2021192023A1 publication Critical patent/WO2021192023A1/ja

Links

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
    • 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
    • G05B19/404—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 control arrangements for compensation, e.g. for backlash, overshoot, tool offset, tool wear, temperature, machine construction errors, load, inertia
    • 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
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37351—Detect vibration, ultrasound
    • 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
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/41—Servomotor, servo controller till figures
    • G05B2219/41121—Eliminating oscillations, hunting motor, actuator

Definitions

  • the present disclosure relates to a numerical control device that controls the operation of a machine tool.
  • the numerical control device that controls the operation of the machine tool causes the machine tool to process the workpiece by operating the servo motor provided by the machine tool in accordance with the command of the machining program.
  • the numerical control device may vibrate the machine tool when operating the machine tool.
  • the machine tool vibrates, if the servomotor that moves the tool vibrates, the positioning accuracy of the tool deteriorates, and the machining accuracy decreases. Therefore, it is desired to reduce the vibration of the machine tool.
  • the compensator described in Patent Document 1 detects a disturbance (vibration) that determines the displacement of the tool from the target amount with respect to the workpiece, and determines the future characteristics of the disturbance based on a mathematical process model formed from the disturbance. It is evaluated and the displacement is corrected according to the future characteristics.
  • a disturbance vibration
  • the present disclosure has been made in view of the above, and an object of the present disclosure is to obtain a numerical control device capable of suppressing aperiodic vibration generated by a command.
  • the numerical control device of the present disclosure gives a position command to a drive unit that drives the tool when the machine tool executes machining on the workpiece by using the tool.
  • a position command to a drive unit that drives the tool when the machine tool executes machining on the workpiece by using the tool.
  • the process corresponding to the auxiliary function is controlled and the first machine tool that occurs when the auxiliary function is executed.
  • It is equipped with a control unit that outputs a suppression position command, which is a position command for suppressing the vibration of the machine tool, to the drive unit.
  • the suppression position command is a position command that causes the drive unit to generate a second vibration having an amplitude equal to or less than the amplitude of the first vibration in the opposite phase to the first vibration.
  • the numerical control device has an effect of being able to suppress aperiodic vibration generated by a command.
  • the figure which shows the structure of the numerical control apparatus which concerns on embodiment Flow chart showing the operation procedure of the numerical control device according to the embodiment
  • the figure for demonstrating the vibration suppressed by the numerical control device which concerns on embodiment The figure which shows the hardware configuration example which realizes the machine learning device provided in the numerical control device which concerns on embodiment.
  • FIG. 1 is a diagram showing a configuration of a numerical control device according to an embodiment.
  • Numerical control (NC) device 1 is a computer that controls a machine tool according to a machining program.
  • the machining program used by the numerical control device 1 to control the machine tool is described by a preparation function called a G code, a feed function called an F code, an auxiliary function called an M code, and the like.
  • the auxiliary function is a function that causes the machine tool to perform operations that are not directly related to machining.
  • auxiliary functions include the function of enabling or disabling the injection of cooling water to suppress the heat generated by friction during machining, the operation of changing tools, the function of automatically opening and closing the doors of machine tools, tables, etc. It is a function to execute clamping or unclamping on the rotating shaft that rotates. Clamping is a process of mechanically locking the rotating shaft, and unclamping is a process of mechanically unlocking the rotating shaft.
  • the table rotated by the rotation axis fixes the work piece with the work piece placed on it.
  • This table can be rotated 360 degrees by rotating the rotation axis.
  • the work piece is rotated to a specific position while being placed on the table.
  • the numerical control device 1 executes the auxiliary function by using the machining program in the numerical control device 1, various operations unique to the machine tool maker can be realized.
  • the auxiliary function is indicated by the auxiliary function number J3, which is the number of the auxiliary function.
  • a machine tool is a device that executes machining on a work piece using a tool.
  • the machine tool includes a drive unit that drives the tool.
  • the components of the machine tool may or may not include tools.
  • the numerical control device 1 outputs a position command J1 for driving the drive unit when the machine tool executes machining on the workpiece. Further, the numerical control device 1 controls the processing corresponding to the auxiliary function when there is a command of the auxiliary function to execute the processing other than the processing of the workpiece. Further, the numerical control device 1 outputs a suppression position command 51, which is a position command for suppressing the vibration of the machine tool generated when the auxiliary function is executed, to the drive unit.
  • the numerical control device 1 is connected to the drive unit 31 and the remote IO (Input / Output) 33.
  • the drive unit 31 is connected to the servo motor 32.
  • the servomotor 32 is an example of a drive unit that drives a tool.
  • the servomotor 32 is connected to a tool post or the like to which a tool for machining an workpiece is attached, and the tool is operated by operating the tool post or the like.
  • the numerical control device 1 includes an analysis unit 21, a control unit 22, a PLC (Programmable Logic Controller) 23, a machine learning device 10, and a machining program storage unit 40.
  • the machining program storage unit 40 stores the machining program used when the machine tool is controlled.
  • the analysis unit 21 analyzes a machining program which is a sequence program.
  • the analysis unit 21 sends the analysis result to the control unit 22.
  • the control unit 22 generates and outputs the position command J1, the selection tool number J2, and the auxiliary function number J3 based on the analysis result sent from the analysis unit 21.
  • the position command J1 is a command that specifies the machining position by the tool.
  • the position command J1 is indicated by, for example, the X coordinate, the Y coordinate, and the Z coordinate.
  • the selected tool number J2 is a number that identifies the tool selected for machining the workpiece.
  • Auxiliary function number J3 is a number that identifies the type of auxiliary function. Since there is a function of generating vibration in a tool or the like among the auxiliary functions, in the present embodiment, the numerical control device 1 suppresses the vibration with respect to the auxiliary function of generating vibration. It is assumed that the auxiliary function number J3 of the auxiliary function for suppressing vibration is set in advance in the numerical control device 1.
  • the setting of the auxiliary function number J3 is set by, for example, a parameter, but the setting method is not limited to the setting by the parameter and may be set by any method.
  • the auxiliary function number J3 is the first identification information for identifying the type of the auxiliary function, and the selected tool number J2 is the second identification information for identifying the type of the tool.
  • the control unit 22 outputs the suppression position command 51 corresponding to the auxiliary function number J3 until the machine learning device 10 learns the suppression position command 51.
  • the suppression position command 51 here is an initial value of the suppression position command 51 registered in advance in the numerical control device 1 for each type of auxiliary function. That is, the control unit 22 outputs the initial value of the suppression position command 51 corresponding to the auxiliary function number J3 before the auxiliary function is first executed. In this way, the control unit 22 outputs the pre-registered suppression position command 51 until the machine learning device 10 learns the suppression position command 51.
  • the suppression position command 51 is a position command for suppressing vibration generated due to the auxiliary function to be executed.
  • the suppression position command 51 is a command that specifies a machining position, which is indicated by, for example, the X coordinate, the Y coordinate, and the Z coordinate.
  • the suppression position command 51 is a position command that generates a vibration (second vibration) having the same magnitude as the vibration of the machine tool (first vibration) in the opposite phase to the drive unit of the machine tool.
  • the suppression position command 51 is a position command capable of suppressing the vibration of the tool.
  • the control unit 22 receives the suppression position command 51 corresponding to the suppression information sent from the machine learning device 10. Is output.
  • the control unit 22 outputs the suppression position command 51 at a timing at which vibration can be suppressed before and after learning the suppression position command 51.
  • the control unit 22 outputs the position command J1 and the suppression position command 51 to the drive unit 31, and outputs the auxiliary function number J3 to the PLC 23. Further, the control unit 22 outputs the position command J1, the selection tool number J2, and the auxiliary function number J3 to the machine learning device 10.
  • the PLC 23 sends an instruction (auxiliary function command) to execute the auxiliary function corresponding to the auxiliary function number J3 sent from the control unit 22 to the remote IO 33.
  • the PLC 23 stores a ladder program in which the machine operation is described.
  • the PLC 23 receives the auxiliary function number J3 indicating the auxiliary function command such as the M code, the PLC 23 sends the auxiliary function command to the remote IO 33 according to the ladder program.
  • the remote IO 33 is a device that transmits / receives input / output signals to and from a device that executes an auxiliary function in the machine tool (hereinafter referred to as an auxiliary processing execution device).
  • the remote IO 33 receives an auxiliary function command indicating an auxiliary function from the numerical control device 1.
  • the remote IO 33 transmits an auxiliary function command to the auxiliary processing execution device, and receives information indicating the status of the auxiliary processing execution device from the auxiliary processing execution device.
  • Examples of auxiliary processing execution devices are a device for injecting cooling water, a device for performing tool change, a device for opening and closing a door provided in a machine tool, and a device for clamping and unclamping a rotating shaft.
  • the auxiliary function is a function executed by the auxiliary function command, and vibration (impact) is generated in the workpiece or the machine tool by the control by the auxiliary function.
  • the vibration (impact) generated by the control by the auxiliary function is aperiodic and cannot be suppressed by the prior art.
  • the drive unit 31 is connected to the servomotor 32 and controls the servomotor 32 according to a command sent from the numerical control device 1.
  • the drive unit 31 receives the position command J1 in which the position of the tool is specified and the suppression position command 51 in which the position of the tool for suppressing vibration is specified from the numerical control device 1.
  • the drive unit 31 controls the position of the tool based on the position command J1.
  • the drive unit 31 controls the position of the tool based on the suppression position command 51.
  • the drive unit 31 calculates the current value to be sent to the servomotor 32 based on the position command J1 and the suppression position command 51 from the numerical control device 1.
  • the drive unit 31 drives the servomotor 32 by sending a current corresponding to the position command J1 and a current corresponding to the suppression position command 51 to the servomotor 32.
  • the servomotor 32 is connected to the turret and rotates the tool attached to the turret according to the current from the drive unit 31.
  • the drive unit 31 receives position information indicating the actual position of the tool from the servomotor 32. Further, the drive unit 31 receives the servomotor temperature J5 indicating the temperature of the servomotor 32 from the servomotor 32.
  • the drive unit 31 transmits the position information indicating the actual position of the tool to the machine learning device 10 as the FB (Feed Back) position J4. Further, the drive unit 31 transmits the temperature of the servomotor 32 to the machine learning device 10 as the servomotor temperature J5.
  • the FB position J4 may be detected by a detector that detects the position of the tool, or may be calculated based on the position command J1 to the servomotor 32.
  • the machine learning device 10 includes a state observation unit 11 and a learning unit 12.
  • the state observation unit 11 acquires the position command J1, the selection tool number J2, and the auxiliary function number J3 from the control unit 22. Further, the state observation unit 11 acquires the FB position J4 and the servomotor temperature J5 from the drive unit 31.
  • the state observation unit 11 determines the position command J1, the selection tool number J2, and the auxiliary.
  • the function number J3, the FB position J4, and the servo motor temperature J5 are acquired as state variables.
  • the suppression information is information for generating a suppression position command 51 for suppressing vibration caused by an auxiliary function.
  • the suppression information includes the amplitude and output timing of the suppression position command 51.
  • the learning unit 12 learns the suppression information corresponding to the auxiliary function based on the state variables obtained from the state observing unit 11.
  • the machine learning device 10 corresponds to the type of tool corresponding to the selected tool number J2 and the auxiliary function. Used for learning suppression information. Further, since the degree of vibration may differ depending on the temperature of the servomotor 32, in the present embodiment, the machine learning device 10 uses the servomotor temperature J5 for learning the suppression information corresponding to the auxiliary function. The numerical control device 1 does not have to use the selected tool number J2 and the servomotor temperature J5 for learning the suppression information.
  • FIG. 2 is a flowchart showing an operation procedure of the numerical control device according to the embodiment.
  • the analysis unit 21 analyzes the machining program (step S10). At this time, the analysis unit 21 determines whether or not the command to be executed next among the commands included in the machining program is the auxiliary function command to be suppressed by vibration (step S20).
  • the control unit 22 sends the vibration suppression target auxiliary function number J3 to PLC23.
  • the PLC 23 transmits the auxiliary function command corresponding to the auxiliary function number J3 to the auxiliary processing execution device via the remote IO 33.
  • the auxiliary processing execution device executes the auxiliary function corresponding to the auxiliary function command.
  • the numerical control device 1 controls the execution of the auxiliary function (step S30).
  • the control unit 22 outputs the suppression position command 51 corresponding to the auxiliary function number J3.
  • the control unit 22 outputs the suppression position command 51 registered in advance as the suppression position command 51 corresponding to the auxiliary function to the drive unit 31.
  • the numerical control device 1 suppresses the vibration caused by the auxiliary function by using the suppression position command 51 corresponding to the auxiliary function.
  • the numerical control device 1 suppresses the vibration of the tool, the tool post, the servomotor 32, etc. by using the suppression position command 51 (step S40). In this way, the numerical control device 1 suppresses vibration caused by the auxiliary function.
  • the control unit 22 When the auxiliary function command is a command to be suppressed by vibration, the control unit 22 notifies the state observation unit 11 of the position command J1, the selected tool number J2, and the auxiliary function number J3. As a result, the state observation unit 11 acquires the position command J1, the selection tool number J2, and the auxiliary function number J3. Further, the state observation unit 11 acquires the FB position J4 and the servomotor temperature J5 of the servomotor 32 from the drive unit 31. The state observation unit 11 sends the position command J1, the selection tool number J2, the auxiliary function number J3, the FB position J4, and the servomotor temperature J5 to the learning unit 12 as state variables.
  • the learning unit 12 learns the suppression information which is the information of the suppression position command 51 corresponding to the auxiliary function based on the state variable sent from the state observation unit 11 (step S50), and controls the learned suppression information. Send to 22. Specifically, the learning unit 12 learns the amplitude and timing of the suppression position command 51 corresponding to the auxiliary function, and sends the suppression information indicating the amplitude and timing of the suppression position command 51 to the control unit 22.
  • step S20 the numerical control device 1 returns to the process of step S20. That is, the analysis unit 21 determines whether or not the command to be executed next is an auxiliary function command for vibration suppression target (step S20).
  • the numerical control device 1 controls the machining process of the workpiece based on the command included in the machining program (step). S60).
  • commands included in the machining program examples of commands other than those subject to vibration suppression are commands related to machining, that is, commands for executing machining on a workpiece using a tool.
  • a command other than the vibration suppression target is a machining command.
  • the control unit 22 When the numerical control device 1 controls the machining process of the workpiece, the control unit 22 generates the position command J1 corresponding to the machining command included in the machining program and sends it to the drive unit 31. As a result, the drive unit 31 controls the servomotor 32 according to the position command J1. After this, the numerical control device 1 returns to the process of step S20. That is, the analysis unit 21 determines whether or not the command to be executed next is an auxiliary function command for vibration suppression target (step S20).
  • the numerical control device 1 After executing the auxiliary function, the numerical control device 1 immediately starts the processing of the workpiece regardless of the presence or absence of vibration of the machine tool. When the numerical control device 1 does not execute the suppression position command, the machine tool executes the machining process in a situation where vibration is generated due to the execution of the auxiliary function. On the other hand, in the present embodiment, the numerical control device 1 outputs the position command J1 corresponding to the machining command while outputting the suppression position command 51, so that the machine tool suppresses the vibration generated by the execution of the auxiliary function. While doing so, the machining process will be executed. The numerical control device 1 may start the machining process of the workpiece when the misalignment of the tool due to vibration becomes smaller than the specific value. Since the numerical control device 1 can suppress the vibration even when the vibration is generated by the execution of the auxiliary function, the time until the vibration is settled can be shortened.
  • step S20 the numerical control device 1 repeats the processes of steps S60 and S20 until the command included in the machining program becomes the auxiliary function command to be suppressed by vibration.
  • step S20 when the command included in the machining program is an auxiliary function command for vibration suppression target (step S20, Yes), the numerical control device 1 controls the execution of the auxiliary function (step S30).
  • the numerical control device 1 suppresses the vibration of the tool, the tool post, the servomotor 32, etc. by using the suppression position command 51 (step S40). Then, the learning unit 12 learns the suppression information corresponding to the auxiliary function based on the state variable sent from the state observing unit 11 (step S50), and the numerical control device 1 returns to the process of step S20.
  • the numerical control device 1 repeats the processes of steps S20 to S50 until all the commands in the machining program are executed.
  • FIG. 3 is a diagram for explaining the vibration suppressed by the numerical control device according to the embodiment.
  • the waveform F1 of the graph shown in the upper part of FIG. 3 shows the change of the FB position J4, and the graph shown in the lower part of FIG. 3 shows the suppression position command 51.
  • the horizontal axis of the graph shown in the upper part of FIG. 3 is time, and the vertical axis is the coordinates of the FB position.
  • the horizontal axis of the graph shown in the lower part of FIG. 3 is time, and the vertical axis is the suppression position command 51 (coordinates of the commanded position).
  • the FB position J4 is indicated by, for example, the X coordinate, the Y coordinate, and the Z coordinate. In the example of FIG. 3, the FB position J4 in the X direction and the suppression position command 51 in the X direction are shown.
  • Waveform F1 is a waveform when the auxiliary function is executed before the machining command is executed. That is, the waveform F1 is a waveform before the movement of the tool with respect to the position command J1 is started. As shown in the waveform F1, the FB position J4 changes when the auxiliary function is executed. That is, vibration is generated by executing the auxiliary function. In order to suppress this vibration, the numerical control device 1 outputs a suppression position command 51 for suppressing the vibration at the timing T1 at which the vibration generated by the execution of the auxiliary function can be suppressed.
  • the suppression position command 51 is a command for suppressing the FB position J4. That is, the suppression position command 51 is a command for setting the FB position J4 (tool position) to 0. Specifically, the suppression position command 51 is a command for moving the tool to a position opposite to the FB position J4, which is the position of the tool. For example, when the FB position J4 has positive coordinates, the suppression position command 51 is a command to move the tool to negative coordinates.
  • the suppression position command 51 has an amplitude A1 capable of suppressing the FB position J4. That is, the position of the tool corresponding to the suppression position command 51 has the same absolute value as the FB position J4 and the sign is opposite.
  • the suppression position command 51 is a position command corresponding to the waveform X1 capable of suppressing the first wave.
  • the waveform X1 has a phase opposite to that of the first wave of the waveform F1.
  • the amplitude of the waveform X1 is preferably the same magnitude as the first wave of the waveform F1, but may be substantially the same as the magnitude of the first wave of the waveform F1. If the amplitude of the waveform X1 is equal to or smaller than the amplitude of the first wave of the waveform F1, vibration can be suppressed.
  • the suppression position command 51 may include a position command capable of suppressing the second and subsequent waves. It is assumed that the suppression position command 51 of the present embodiment includes a position command up to the Nth (N is a natural number) wave so that the position deviation of the tool due to vibration becomes smaller than a specific value.
  • the numerical control device 1 starts the machining process of the workpiece by using the position command J1 when the misalignment of the tool due to vibration becomes smaller than the specific value. As a result, the FB position J4 also becomes a position corresponding to the position command J1.
  • the suppression position command 51 is generated based on the timing T1 for generating the command and the amplitude A1 of the command included in the suppression information. Then, the vibration due to the auxiliary function is suppressed by the generated suppression position command 51.
  • an example of the starting point of the timing T1 for generating the suppression position command 51 is the time when the target auxiliary function is notified from the analysis unit 21 to the control unit 22.
  • the starting point of the timing T1 for generating the suppression position command 51 may be any timing as long as the vibration can be suppressed.
  • the state observation unit 11 observes the position command J1, the selection tool number J2, the auxiliary function number J3, the FB position J4, and the servomotor temperature J5 as state variables.
  • the learning unit 12 learns the amplitude and timing of the suppression position command 51, which is suppression information, according to the data set created based on the state variables.
  • the numerical control device 1 may use any learning algorithm used for the learning unit 12.
  • the learning algorithm a case where reinforcement learning (Reinforcement Learning) is applied will be described.
  • an agent action subject
  • an agent action subject
  • the current state environment parameter
  • the environment changes dynamically depending on the behavior of the agent, and the agent is rewarded according to the change in the environment.
  • the agent repeats this process and learns the action policy that gives the most reward through a series of actions.
  • agents are rewarded by the environment by choosing an action and learn how to get the most reward through a series of actions.
  • Q-learning and TD-learning are known as typical methods of reinforcement learning.
  • the general update equation (learned model) of the action value function Q (s, a) is expressed by the following equation (1).
  • s t represents the environment at time t
  • a t represents the behavior in time t.
  • state environment
  • rt + 1 represents the reward received by the change of the state
  • ⁇ represents the discount rate
  • ⁇ represents the learning coefficient. Note that ⁇ is in the range of 0 ⁇ ⁇ 1 and ⁇ is in the range of 0 ⁇ ⁇ 1.
  • inhibition data (amplitude and timing of the suppression position command 51) act a t next position command J1, is selected tool number J2, auxiliary function number J3, FB position J4, and the servo motor temperature J5 state s t, and the learning section 12 learns the best action a t in state s t at time t.
  • the learning unit 12 updates the action value function Q (s, a) so that the action value Q of the action a at the time t approaches the best action value Q at the time t + 1.
  • the best behavioral value Q in a certain environment is sequentially propagated to the behavioral value Q in the previous environment.
  • the learning unit 12 includes a reward calculation unit 15 and a function update unit 16.
  • the reward calculation unit 15 calculates the reward r based on the state variable.
  • the reward calculation unit 15 calculates the reward r based on the difference between the position command J1 and the FB position J4, that is, the error due to vibration. For example, the reward calculation unit 15 increases the reward r (for example, gives a reward of "1") when the difference between the position command J1 and the FB position J4 is small.
  • the reward calculation unit 15 reduces the reward r (for example, gives a reward of "-1") when the difference between the position command J1 and the FB position J4 is large.
  • the learning unit 12 sets the difference between the position command J1 and the FB position J4 as the total difference in the time from the auxiliary function command to the next command.
  • the learning unit 12 may determine the difference acquisition time by a parameter or the like, and use the total from the auxiliary function command to the determined acquisition time as the difference.
  • the function update unit 16 updates the function for determining the suppression information according to the reward r calculated by the reward calculation unit 15. For example, in the case of Q learning function updating unit 16, action value represented by the formula (1) function Q (s t, a t) and is used as a function for calculating a suppression information.
  • the learning unit 12 repeatedly executes the above learning.
  • the learning algorithm is not limited to reinforcement learning.
  • the learning algorithm in addition to reinforcement learning, supervised learning, unsupervised learning, semi-supervised learning, and the like can also be applied.
  • the learning unit 12 can also use deep learning, which learns the extraction of the feature amount itself, and other known methods such as a neural network, genetic programming, and a function.
  • Machine learning may be performed according to logical programming, support vector machines, and the like.
  • the machine learning device 10 is used to learn the suppression information of the numerical control device 1, and the machine learning device 10 is connected to the numerical control device 1 via a network, for example. It may be a device separate from 1. Further, the machine learning device 10 may be built in the numerical control device 1. Further, the machine learning device 10 may exist on the cloud server.
  • the learning unit 12 may learn the suppression information according to the data sets created for the plurality of numerical control devices 1.
  • the learning unit 12 may acquire data sets from a plurality of numerical control devices 1 and a plurality of drive units 31 used at the same site, or a plurality of machine tools operating independently at different sites. And the suppression information may be learned by using the data set collected from the plurality of drive units 31. Further, the learning unit 12 can add the numerical control device 1 for collecting the data set to the target on the way, or conversely remove it from the target. Further, the learning unit 12 attaches the machine learning device 10 that has learned the suppression information with respect to a certain numerical control device 1 to another numerical control device 1, and regenerates the suppression information with respect to the other numerical control device 1. You may learn and update.
  • control unit 22 When the control unit 22 receives the auxiliary function command from the analysis unit 21, it determines whether or not the machine learning device 10 is a learned auxiliary function. If the auxiliary function has already been learned by the machine learning device 10, the control unit 22 generates the suppression position command 51 based on the suppression information corresponding to the auxiliary function. The numerical control device 1 suppresses the vibration of the machine tool by using the generated suppression position command 51 when executing the auxiliary function.
  • the numerical control device 1 sets the suppression information in a parameter or the like, creates a suppression position command 51 based on the parameter, and vibrates. May be suppressed.
  • FIG. 4 is a diagram showing an example of a hardware configuration that realizes a machine learning device included in the numerical control device according to the embodiment.
  • the machine learning device 10 can be realized by an input device 300, a processor 100, a memory 200, and an output device 400.
  • An example of the processor 100 is a CPU (Central Processing Unit, central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, DSP (Digital Signal Processor)) or system LSI (Large Scale Integration).
  • Examples of the memory 200 are RAM (Random Access Memory) and ROM (Read Only Memory).
  • the machine learning device 10 is realized by the processor 100 reading and executing a computer-executable learning program for executing the operation of the machine learning device 10 stored in the memory 200. It can be said that the learning program, which is a program for executing the operation of the machine learning device 10, causes the computer to execute the procedure or method of the machine learning device 10.
  • the learning program executed by the machine learning device 10 has a modular configuration including a state observation unit 11 and a learning unit 12, and these are loaded on the main storage device and these are generated on the main storage device. NS.
  • the input device 300 receives the position command J1, the selection tool number J2, the auxiliary function number J3, the FB position J4, and the servomotor temperature J5 and sends them to the processor 100.
  • the memory 200 is used as a temporary memory when the processor 100 executes various processes.
  • the memory 200 stores a state variable, a reward, an action value function, and the like.
  • the output device 400 outputs the suppression information corresponding to the auxiliary function to the control unit 22.
  • the learning program is a file in an installable format or an executable format, and may be stored in a computer-readable storage medium and provided as a computer program product. Further, the learning program may be provided to the machine learning device 10 via a network such as the Internet. Some of the functions of the machine learning device 10 may be realized by dedicated hardware such as a dedicated circuit, and some may be realized by software or firmware. Further, the numerical control device 1 can also be realized by the same hardware configuration as the machine learning device 10.
  • the numerical control device 1 of the present embodiment can suppress the vibration of the machine tool even if the pressure of the clamp is increased to shorten the operation time for the clamp or the unclamp.
  • the numerical control device 1 of the embodiment when the numerical control device 1 of the embodiment has an auxiliary function command, it controls the processing corresponding to the auxiliary function and suppresses the vibration generated when the auxiliary function is executed.
  • 51 is output to the servomotor 32 at the timing of suppressing vibration.
  • the suppression position command 51 is a position command that causes the servomotor 32 to generate vibration having the same magnitude of amplitude in the opposite phase to the vibration of the machine tool.
  • the numerical control device 1 can suppress the aperiodic vibration generated by the command when the auxiliary function is executed.
  • the numerical control device 1 can suppress the vibration of the servomotor 32, the machining accuracy of machining by the machine tool is improved. Further, since the numerical control device 1 can suppress vibration, it is possible to prevent the occurrence of a servo alarm. Further, since the numerical control device 1 can suppress the vibration caused by the operation of the auxiliary function, the operation of the auxiliary function operation can be speeded up and the machining time can be shortened.
  • the configuration shown in the above embodiment is an example, and can be combined with another known technique, or a part of the configuration may be omitted or changed without departing from the gist. It is possible.

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Manufacturing & Machinery (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Numerical Control (AREA)
PCT/JP2020/012983 2020-03-24 2020-03-24 数値制御装置 WO2021192023A1 (ja)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2020542180A JP6775720B1 (ja) 2020-03-24 2020-03-24 数値制御装置
CN202080098600.XA CN115315665A (zh) 2020-03-24 2020-03-24 数控装置
DE112020006576.6T DE112020006576T5 (de) 2020-03-24 2020-03-24 Numerisches Steuergerät
PCT/JP2020/012983 WO2021192023A1 (ja) 2020-03-24 2020-03-24 数値制御装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/012983 WO2021192023A1 (ja) 2020-03-24 2020-03-24 数値制御装置

Publications (1)

Publication Number Publication Date
WO2021192023A1 true WO2021192023A1 (ja) 2021-09-30

Family

ID=72916077

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/012983 WO2021192023A1 (ja) 2020-03-24 2020-03-24 数値制御装置

Country Status (4)

Country Link
JP (1) JP6775720B1 (de)
CN (1) CN115315665A (de)
DE (1) DE112020006576T5 (de)
WO (1) WO2021192023A1 (de)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117075535A (zh) * 2023-10-16 2023-11-17 上海诺倬力机电科技有限公司 振动抑制方法、装置、电子设备及可读存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE112022000207T5 (de) * 2021-01-14 2023-09-07 Fanuc Corporation Maschinelle Lernvorrichtung
WO2023223470A1 (ja) * 2022-05-18 2023-11-23 三菱電機株式会社 制御パラメータ調整装置、数値制御装置および制御パラメータ調整方法
JP7490149B1 (ja) 2023-05-10 2024-05-24 三菱電機株式会社 消費電力量調整装置、数値制御装置、および消費電力量調整方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000505578A (ja) * 1996-03-01 2000-05-09 フラウンホーファー・ゲゼルシャフト ツア フェルデルンク デア アンゲヴァンテン フォルシュンク アインゲトラーゲナー フェアアイン 切削工作機械における動的変位を補正する方法および装置
JP2004268177A (ja) * 2003-03-06 2004-09-30 Fanuc Ltd 工具の切粉付着防止／除去方法、及び工具交換装置
JP2006155351A (ja) * 2004-11-30 2006-06-15 Fanuc Ltd 制御装置
JP2019153085A (ja) * 2018-03-02 2019-09-12 ブラザー工業株式会社 振動抑制装置と振動抑制方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS54113779A (en) * 1978-02-24 1979-09-05 Toyo Electric Mfg Co Ltd Digital servo-device
JP6693013B2 (ja) * 2016-02-04 2020-05-13 ブラザー工業株式会社 振動抑制装置と振動抑制方法
JP6903485B2 (ja) * 2017-05-12 2021-07-14 株式会社日立製作所 制振装置および加工機
DE112018000172B4 (de) * 2018-07-05 2020-09-24 Mitsubishi Electric Corporation Numerische Steuervorrichtung
WO2020208685A1 (ja) * 2019-04-08 2020-10-15 三菱電機株式会社 数値制御装置およびびびり振動の発生判定方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000505578A (ja) * 1996-03-01 2000-05-09 フラウンホーファー・ゲゼルシャフト ツア フェルデルンク デア アンゲヴァンテン フォルシュンク アインゲトラーゲナー フェアアイン 切削工作機械における動的変位を補正する方法および装置
JP2004268177A (ja) * 2003-03-06 2004-09-30 Fanuc Ltd 工具の切粉付着防止／除去方法、及び工具交換装置
JP2006155351A (ja) * 2004-11-30 2006-06-15 Fanuc Ltd 制御装置
JP2019153085A (ja) * 2018-03-02 2019-09-12 ブラザー工業株式会社 振動抑制装置と振動抑制方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117075535A (zh) * 2023-10-16 2023-11-17 上海诺倬力机电科技有限公司 振动抑制方法、装置、电子设备及可读存储介质
CN117075535B (zh) * 2023-10-16 2024-01-02 上海诺倬力机电科技有限公司 振动抑制方法、装置、电子设备及可读存储介质

Also Published As

Publication number Publication date
JPWO2021192023A1 (de) 2021-09-30
JP6775720B1 (ja) 2020-10-28
DE112020006576T5 (de) 2022-12-01
CN115315665A (zh) 2022-11-08

Similar Documents

Publication Publication Date Title
WO2021192023A1 (ja) 数値制御装置
JP6542839B2 (ja) 制御装置及び機械学習装置
US11000949B2 (en) Robot for controlling learning in view of operation in production line, and method of controlling the same
JP6494897B1 (ja) 数値制御装置
US20220308531A1 (en) Motor controller and motor control method
JP2018118353A (ja) 学習制御機能を備えた制御システム及び制御方法
JP6386516B2 (ja) 学習機能を備えたロボット装置
JP6811908B1 (ja) 数値制御装置、機械学習装置および数値制御方法
CN113168155B (zh) 数控装置及数控方法
US10691091B2 (en) Controller and machine learning device
JP2020035159A (ja) パラメータ調整装置
WO2021186574A1 (ja) 制御システム、モータ制御装置および機械学習装置
WO2020162199A1 (ja) 制御装置、モデル作成方法および制御プログラム
US11022951B2 (en) Information processing device and information processing method
CN112867975B (zh) 维护辅助***、数控装置及维护辅助***的控制方法
CN112975004A (zh) 优劣判定装置及优劣判定方法
JP2020055095A (ja) 制御装置及び制御システム
JP2001154719A (ja) 自由曲線補間方法
JP6800384B1 (ja) 位置決め制御装置および位置決め方法
JP4200357B2 (ja) 電動機制御装置の運転立上げ方法および電動機制御装置
WO2020194685A1 (ja) 数値制御装置および機械学習装置
CN117813561A (zh) 运动控制方法及装置

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2020542180

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20926415

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20926415

Country of ref document: EP

Kind code of ref document: A1