CN118265892A - Takt time division data collection system and abnormality detection system - Google Patents

Abstract

The tact divided data collection system includes a processor. The processor of the takt time division data collection system performs processing for determining whether or not the trigger waveform is included in the takt time waveform in the time-series data acquired from the sensor, based on the trigger waveform stored in the storage device in advance as a part of the takt time waveform of the repeating waveform. When it is determined that the takt waveform includes a trigger waveform, the processor of the takt divided data collection system collects data of the takt waveform including the trigger waveform.

Description

Takt time division data collection system and abnormality detection system

Technical Field

The present invention relates to a tact-divided data collection system and an abnormality detection system.

Background

The frequency of failure varies greatly depending on the number of years after the industrial equipment is introduced. Statistically, the initial failure period of failure due to design and manufacturing defects 1 to 2 years after the introduction, the occasional failure period of failure which is difficult to cause failure and stable operation about 3 to 6 years after the introduction, and the wear failure period of failure increase due to the life of the machine 7 to 8 years after the introduction are shown. Although the initial failure period is supported by the initial failure handling at the manufacturer side, the wear failure period is difficult to read depending on the use condition of the machine, the maintenance frequency of the user, and the like, and if the state at the user side is not grasped, there is a problem that the operation rate is lowered and productivity is easily lowered. With respect to the devices in this period, since the latest monitoring solutions on the manufacturer side are often difficult to be introduced due to old models, there is a tendency that the need for solutions for implementing abnormality detection by newly adding and monitoring a general-purpose sensor increases. Such additional sensors include a temperature sensor and a vibration sensor, and in the case of a device driven by a motor, a current sensor for measuring a motor drive current is used. In the case where the target equipment is a processing machine for processing a workpiece unit such as machining or a vertical multi-joint robot for transporting a workpiece, since the measurement data is a repetition of "tact" indicating a series of steps, a means for detecting a difference by comparing the repeated tacts in abnormality detection is generally used.

Patent document 1 proposes a method of performing pattern matching between a predetermined tact reference waveform and acquired data to realize tact separation.

Prior art literature

Patent literature

Patent document 1: japanese patent laid-open No. 2020-187502

Disclosure of Invention

Problems to be solved by the invention

In the case of abnormality detection by the additional sensor, it is desirable to minimize the influence on the conventional device, and therefore it is often difficult to acquire a trigger signal for the start of the tact from the control system. Since there are many cases where a deviation occurs between tacts due to waiting of a workpiece, waiting of work of peripheral equipment, or the like, tact division may be mistaken in fixed cycle cutting at tact length, and it is a problem how to realize tact division with high accuracy without a trigger signal.

In order to meet the above requirements, in the system described in patent document 1, a learning stage is provided before operation, a tact reference waveform is determined from data acquired in advance and stored, pattern matching between the data acquired so far and the tact reference waveform is performed at regular intervals during operation, it is determined that tact is present when the matching degree exceeds a certain amount, and processing for cutting out the data as tact data is performed. There are two disadvantages in this approach. First, since pattern matching is performed on the entire tact reference waveform, it is considered that the calculation load is high and it is not easy to simultaneously install the pattern matching in the data recording apparatus. Second, since the cut acquired data is processed together when matching is established, if the data is to be stored in the system, the I/O load of the data is large, and in this point, the data recording is highly likely to be affected, and therefore, it is considered that the data recording device is not easy to install at the same time. In view of the above two drawbacks, it is necessary to prepare a device for dividing the tact separately from the data recording device, and there is a problem that the cost of the device increases.

The present invention has been made in view of the above-described points, and an object thereof is to realize takt cutting while recording, and to perform takt division data collection at low cost while suppressing the number of devices.

Means for solving the problems

According to a first aspect of the present invention, the following tact split data collection system is provided. That is, the tact divided data collection system includes a processor. The processor determines whether or not the trigger waveform is included in the time-series data acquired from the sensor, based on the trigger waveform stored in the storage device in advance as a part of the tact waveform of the repeated waveform. When the processor determines that the takt waveform includes a trigger waveform, the processor collects data of the takt waveform including the trigger waveform.

According to a second aspect of the present invention, the following abnormality detection system is provided. That is, the abnormality detection system includes a tact divided data collection module and an abnormality detection module. The takt dividing data collection module includes a sensor data acquisition unit, a takt waveform storage unit, a trigger waveform storage unit, a takt start determination unit, and a takt data storage unit. The sensor data acquisition unit acquires time-series data from the sensor. The tact waveform storage section stores one of the repeated waveforms included in the learning-time data as a tact reference waveform. The trigger waveform storage section stores a waveform of a fixed amount of time from the beginning of the tact reference waveform as a trigger waveform that is a tact start trigger. The tact start determination section includes a process of extracting data of a trigger waveform length amount from the data at the time of operation and calculating a similarity with the trigger waveform. The tact data storage unit stores the operation time data as a file for each tact length based on the calculation result. The abnormality detection module makes a normal/abnormal determination using the obtained data for each tact.

Effects of the invention

According to the present invention, there is provided a system capable of realizing tact cutting while recording and conducting tact division data collection at low cost with the number of devices suppressed. The problems, structures, and effects other than those described above will be apparent from the following description of the embodiments for carrying out the invention.

Drawings

Fig. 1 is a diagram showing the external appearance of a data collection system and a target device according to a conventional example.

Fig. 2 is a diagram showing a power supply system circuit of a device for a data collection system according to a conventional example.

Fig. 3A is a diagram showing a current waveform obtained by sensing a device for a data collection system according to the conventional example and the present embodiment, and is a diagram showing an example of a current waveform including a plurality of tacts.

Fig. 3B is a diagram showing a current waveform obtained by sensing a device for a data collection system according to the conventional example and the present embodiment, and is a diagram showing an example of a reference waveform showing 1 tact.

Fig. 4 is a flowchart showing a flow of processing of the takt time division data collection system of the related art.

Fig. 5A is a block diagram of a takt time division data collection system of the conventional example, which is a block diagram in learning.

Fig. 5B is a block diagram of a takt time division data collection system according to a conventional example, and is a block diagram during operation.

Fig. 6 is a diagram showing an example of a formula of statistics.

Fig. 7 is a diagram showing an external appearance of an example of the takt time division data collection system and the target device according to the embodiment.

Fig. 8 is a diagram showing an example of a power supply system circuit of the device for which the data collection system according to the embodiment is intended.

Fig. 9 is a diagram showing an example of a trigger waveform of the start of acquisition of 1 takt of a current waveform obtained by sensing a target device by the takt time division data collection system according to the first embodiment.

Fig. 10 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the first embodiment.

Fig. 11A is a block diagram showing an example of processing of the takt time division data collection system according to the embodiment, and shows a block diagram during learning.

Fig. 11B is a block diagram showing an example of processing of the takt time division data collection system according to the embodiment, and shows a block diagram during operation.

Fig. 12 is a diagram showing an example of a trigger waveform of the start of acquisition of 1 takt of a current waveform obtained by sensing a target device by the takt time division data collection system according to the first embodiment.

Fig. 13 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the second embodiment.

Fig. 14 is a diagram showing an example of the formula of NCC.

Fig. 15 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the third embodiment.

Fig. 16 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the fourth embodiment.

Fig. 17 is a diagram showing an example of the expression of the autocorrelation function.

Fig. 18A is a diagram showing a current waveform obtained by sensing a target device by the takt time division data collection system according to the fifth embodiment, and is a diagram showing an example of a current raw waveform.

Fig. 18B is a diagram showing a current waveform obtained by sensing a target device by the takt time division data collection system according to the fifth embodiment, and is a diagram showing an example of a moving RMS waveform.

Fig. 19 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the fifth embodiment.

Fig. 20 is a flowchart showing an example of the processing flow of the tact divided data collection system according to the sixth embodiment.

Fig. 21 is a diagram showing an external appearance of an example of an abnormality detection system and an object device according to the seventh embodiment.

Fig. 22 is a flowchart showing an example of the processing flow of the abnormality detection system according to the seventh embodiment.

Fig. 23A is a block diagram showing an example of processing of the abnormality detection system according to the seventh embodiment, and is a block diagram during learning.

Fig. 23B is a block diagram showing an example of processing of the abnormality detection system according to the seventh embodiment, and is a block diagram during operation.

Fig. 24 is a hardware configuration diagram relating to an example of a system configuration.

Detailed Description

As described above, in the takt divided data collection system using the configuration in which pattern matching is performed using the reference waveform described in patent document 1, the data can be stored together when pattern matching is performed using the obtained data of the takt length amount and the takt reference waveform. In such a method, as described above, the calculation load and the I/O load are high, and the recording apparatus cannot be mounted at the same time, and there is a possibility that the cost increases due to the addition of the apparatus.

First, an example of a device and a sensor data structure for a takt time division data collection system according to the conventional example will be described.

Fig. 1 is a diagram showing an example of a conventional takt time division data collection system and a target apparatus, which is exemplified by a robot. In this system, the robot is configured to include a control box 200 and a robot part 300. In the robot part 300, there are a plurality of operation shafts, and the first operation shaft 301, the second operation shaft 302, the third operation shaft 303, the fourth operation shaft 304, the fifth operation shaft 305, and the sixth operation shaft 306 are capable of rotational movement in the arrow directions, respectively.

The tact divided data collection system 1 is connected to the control box 200 of the subject device. The tact divided data collection system 1 includes a storage unit 10 and a processing unit 20. The storage unit 10 stores a tact reference waveform storage unit 11 for storing a reference waveform of tact and a tact data storage unit 13 for storing divided data. The processing unit 20 includes a sensor data acquisition unit 21 and a tact start determination unit 22.

Fig. 2 is a diagram showing an example of a sensing target circuit of a target device of the tact split data collection system. Power is supplied from the control box 200 to each motor, which is a driving motor in the robot 300. The robot 300 includes a plurality of drive motors, and the rotation of each drive motor is transmitted to a speed reducer to operate an operation shaft provided at the tip thereof. The speed reducer is a mechanism that reduces the input rotational speed and increases the torque instead.

In the example of fig. 2, the control box 200 includes a power supply circuit 210 and a control board 220, and electric power supplied from the switchboard 100 is delivered to the power supply circuit 210 and supplied to the respective motors through the control board 220. The control board 220 can supply electric power to a plurality of electric power consuming devices. As the driving motors, a first motor 321, a second motor 322, a third motor 323, a fourth motor 324, a fifth motor 325, and a sixth motor 326 are mounted on the robot portion 300.

The first decelerator 311, the second decelerator 312, the third decelerator 313, the fourth decelerator 314, the fifth decelerator 315, and the sixth decelerator 316 are connected to the respective driving motors. In addition, a first operation shaft 301, a second operation shaft 302, a third operation shaft 303, a fourth operation shaft 304, a fifth operation shaft 305, and a sixth operation shaft 306 are connected to each of the reducers, respectively. In the robot 300, the motors connected to the respective operation axes are driven by the magnitude and frequency of the current output from the control board 220 according to the operation program. Therefore, these current waveforms are known to be important physical quantities reflecting the degradation states of the motors of the respective operation axes of the robot, the speed reducers, which are mechanical components connected to the motors, and the operation axes. In the conventional example, the current sensor is sandwiched in a power supply circuit connected to the first motor therein to perform sensing.

Fig. 3 is a diagram schematically showing a waveform of a current obtained by the current sensor, fig. 3A is a diagram showing an example of a current waveform including a plurality of tacts, and fig. 3B is a diagram showing an example of a reference waveform showing 1 tact. The 1 tact represents the minimum unit of a series of repetitive operations of the robot, and in fig. 3A, the tact divided data collection system is a system that acquires continuous data and outputs the 1 tact amount data of fig. 3B as continuous data including 4 tact amounts. Fig. 4 is a diagram showing a flow of processing of the tact-divided-data collection system of the conventional example, and fig. 5 is a block diagram of the tact-divided-data collection system of the conventional example.

A takt-dividing data collection system according to the conventional example will be described first with reference to fig. 1 to 5, based on the procedure at the time of learning. As described above, the current sensor is sandwiched between the first motor for operating the first operation shaft 301 of the robot unit 300 of fig. 1 and 2 and the wiring between the control board 220, and continuous current waveforms connected to each other by the plurality of tacts of fig. 3A are obtained. This flow is a sequence shown in item numbers 0, 1 (S1, S2) of fig. 4, and corresponds to the block diagram at the time of learning of fig. 5A. The tact-divided-data collection system 1 observes the continuous current waveform, determines a repetitive tact waveform, and stores it as a tact reference waveform.

The sensor data acquisition unit 21 collects and stores sensor data of a predetermined amount of time of 100msec to 1sec as a file, and stores data of 1sec as one temp file (temporary file) in this example, and sets the takt length as 60sec in this example.

In the step of determining in the tact reference waveform storage unit 11, the stored 1-sec files are analyzed by connecting, for example, 3600 files for 1 hour, and a tact for 60sec is clarified to be present, and a job of storing a tact reference waveform for 60sec is performed. The steps at the time of use will be described. In use, as shown in item No. 2 (S3) of fig. 4, 60 temp files are acquired, combined and held, and a step of comparing the acquired files with the reference waveforms of item nos. 3 and 4 (S4 and S5) is prepared. In the conventional example, the comparison means uses statistics represented by the equation of fig. 6.

Here, RMS represents Root Mean Square (Root Mean Square), n represents the data amount of the tact length amount, x _i represents the sensor value, i.e., the current value, and one value is calculated in the entire tact length waveform. In item No. 3, RMS of the tact reference waveform determined and held in advance and RMS of the acquired combination data of 60 seconds obtained during operation are calculated, respectively. In item No. 4, it is determined whether or not the absolute value of the difference of RMS is equal to or less than a predetermined value, for example, when the RMS value of the tact reference waveform is 8[ arms ], and the RMS value of the obtained combination data is 8.5[ arms ], the absolute value of the difference of RMS is 0.5[ arms ].

Here, if the threshold value is 0.3[ arms ], the absolute value of the difference in RMS becomes equal to or greater than the threshold value, and the process shifts to item number 5 (S6). In item No. 5, since the acquired connection data does not match the tact reference waveform, a job is performed in which one file is deleted from the old party in order to update the acquired connection data. Thus, when the item number 2 is returned, a new file is added, and the process shifts to the tact start determination step again.

If the absolute value of the difference in RMS is equal to or less than the threshold value, the process shifts to item number 6 (S7) because the cycle of item numbers 2 to 5 is repeated and the tact is considered to be present in the acquired combination data. In item No. 6, the acquired bond data is stored as a tact file, and after the acquired bond data is stored in the tact data storage unit, the operation of item No. 7 (S8) is performed in which all temp files related to the acquired bond data are deleted. Since the tact does not substantially overlap with the next tact, the data is temporarily reset like this, waiting for the next tact. In addition, in item No. 0 (S1), the processing unit 20 executes the sensor data acquisition unit 21, and in item nos. 1 to 7 (S2 to S8), the processing unit 20 executes the tact start determination unit 22.

With the above-described flow, in the conventional example, the tact divided data collection system can be constructed, but in item No. 3, since the statistical value calculation process of the tact time amount is performed, there is a possibility that the sensor data acquisition operation and the like that are processed in real time are affected. Therefore, as shown by the dotted arrow of fig. 4, there is a disadvantage in that the equipment cost increases due to the separation of the equipment to be processed. Further, since the data of the tact amount needs to be stored together in the storage of the tact amount holding data of item No. 6, there is a possibility that the real-time sensor data acquisition process is affected from the viewpoint of I/O, and there is a possibility that an increase in equipment cost may occur.

Next, an embodiment for solving such problems will be described. The embodiments are examples for explaining the present invention, and are omitted and simplified as appropriate for clarity of explanation. The invention can also be implemented in other various ways. Each component may be single or plural, as long as it is not particularly limited. For easy understanding of the invention, the positions, sizes, shapes, ranges, and the like of the respective constituent elements shown in the drawings may not indicate actual positions, sizes, shapes, ranges, and the like. Accordingly, the present invention is not necessarily limited to the position, size, shape, scope, etc. disclosed in the accompanying drawings.

In the following embodiments, the method capable of suppressing the calculation load and the I/O load has an effect of realizing a low-cost tact split data collection system. Moreover, from the viewpoint of realizing low cost, resource saving can be realized.

In the following embodiments, the same reference numerals denote the same structures even when the figures are different, and the same operational effects are obtained. In addition, the descriptions of the already described contents and the same contents may be omitted.

< First embodiment >, first embodiment

The takt time division data collection system in the first embodiment is configured as shown in fig. 7, and as an example, as shown in fig. 8, the power supply configuration and the sensor installation portion of the target device can be the same as those of the conventional example described using fig. 2. That is, the tact divided data collection system 3 includes a storage unit 30 and a processing unit 40. The target device may be a robot, which includes the control box 500 and the robot unit 600. The control box 500 includes a power supply circuit 510 and a control board 520, and electric power supplied from the switchboard 400 is delivered to the power supply circuit 510 and supplied to the respective motors through the control board 520. The robot section 600 includes motors (621 to 626), decelerator (611 to 616), and operation shafts (601 to 606). Fig. 9 is a diagram showing an example of a trigger waveform of the first embodiment, fig. 10 is a flowchart showing an example of a processing flow of the takt time division data collection system of the first embodiment, and fig. 11 is a block diagram showing an example of a processing of the takt time division data collection system of the first embodiment.

Hereinafter, a common feature in the embodiment of the present invention is the processing of the double frame lines described in item numbers 2,4, and 7 (S12, S14, and S17) in the flowchart of fig. 10. Here, in learning, after the tact reference waveform is determined, the trigger waveform is determined.

The trigger waveform is data obtained by cutting a part of the tact reference waveform, and is used for determining whether or not tact is present in the data amount equal to or less than the tact length. As shown in fig. 11A, the reference waveforms to the tact time are determined in the same order as in the conventional example, and the reference waveforms to the tact time are stored in the reference waveform storage section 31.

Then, using this data, the unique waveform as shown in fig. 9, which exists in the tact and does not exist repeatedly in the tact, is determined in the trigger waveform storage section 32. Here, the unique waveform is, for example, the first 10 seconds. After the determination, the unique waveform is stored in the trigger waveform storage unit 32 as a trigger waveform. This sequence corresponds to the steps of item numbers 0 to 2 (S10 to S12) of fig. 10.

In addition, at the time of use, as shown in item number 3 (S13) of fig. 10, only 10sec data of the trigger amount is acquired, combined, and held. In the next item No. 4 (S14), the same statistic calculation and comparison as in the conventional example are performed from the trigger waveform length data. When the absolute value of the statistical difference is equal to or smaller than the threshold in item number 5 (S15), the process proceeds to item number 7 (S17). Item No. 7 (S17) adds the acquired data to the trigger waveform amount data, and sequentially acquires the remaining data of the tact length amount from the file. Then, when the tact length is reached, the tact file is stored as a tact file as described in item 8 (S18). In addition, all temp files combined in item number 9 (S19) are deleted.

In the present embodiment, as shown in fig. 11B, the tact start determination calculation can be performed with data of a trigger waveform length amount shorter than the tact length. For example, as described above, when the takt length is 60sec and the trigger waveform length is 10sec, the takt start determination calculation can be set to 1/6 as compared with the conventional example, and the calculation cost of the takt divided data collection system 3 can be reduced.

As shown in fig. 12, the intermediate portion of the tact reference waveform may be used as the trigger waveform, and in this case, data from the beginning to the start of the trigger waveform needs to be temporarily held together with the trigger waveform amount data. Since only the trigger waveform length data is used for the takt start determination calculation, the same calculation cost reduction effect can be obtained, but if the memory cost held temporarily is considered, the memory cost increase can be suppressed by using the unique waveform from the beginning as the trigger waveform. In addition, from the same point of view, a trigger waveform of the terminal portion of the tact reference waveform may also be used.

In the first embodiment, as an example, the processing unit 40 executes the sensor data acquisition unit 41 in item number 0 (S10), and the processing unit 40 executes the tact start determination unit 42 in item numbers 1 to 9 (S11 to S19).

< Second embodiment >

Fig. 13 is a flowchart showing a flow of processing of the takt time division data collection system of the second embodiment, which is similar to the configuration shown in fig. 7, and the power supply configuration and the sensor installation site of the target device are similar to those described with reference to fig. 8. In addition, the block diagram of the process of the tact divided data collection system of the second embodiment is the same as fig. 11.

Hereinafter, a common feature of the embodiments of the present invention is that the temporary storage form of the data described by item number 0 in the flowchart of fig. 13 and the Normalized Cross Correlation (NCC) are used for the tact start judgment described by item numbers 4 and 5.

In the temporary storage form of data, the first embodiment is file storage, whereas the second embodiment is temporary storage into a memory. This suppresses the I/O load when the acquired data is stored. But because additional memory storage area is required, the structures need to be selected according to the device used. The main effect in the second embodiment is achieved by using NCC as another feature. Here, NCC is calculated from the equation of fig. 14.

Here, n represents the data amount of the trigger waveform length amount, x _i represents the acquired trigger waveform length amount data, and y _i represents a predetermined trigger waveform. NCC is an index used when evaluating similarity between 2 waveforms. By using NCC, compared with the case of using statistics as shown in the first embodiment, there are a point that the similarity of the entire waveform can be evaluated and a point that the threshold can be determined by a relative value such as the similarity. For example, in NCC, the similarity of 2 waveforms can be calculated with a value between 0 and 1, and therefore, the effect of reducing the SI man-hour of selecting a threshold value for each object can be obtained.

According to this configuration, the accuracy of the tact start determination can be improved, and the number of threshold setting steps for tact start determination can be reduced. In the present embodiment, the determination method using NCC is described, but a difference Square Sum (SSD) and a difference absolute Sum (SAD) which are other typical methods of calculating the similarity between waveforms may be used, and the accuracy of the takt start determination can be improved as in the case of using NCC.

In the second embodiment, as an example, the processing unit 40 executes the sensor data acquisition unit 41 in item number 0 (S20), and the processing unit 40 executes the tact start determination unit 42 in item numbers 1 to 9 (S21 to S29).

< Third embodiment >

Fig. 15 is a flowchart showing an example of a flow of processing of the takt time division data collection system of the third embodiment, in which the takt time division data collection system of the third embodiment is similar to the configuration shown in fig. 7, and the power supply configuration and the sensor installation site of the target device are similar to those of the example described using fig. 8. In addition, the block diagram of the process of the tact divided data collection system of the third embodiment is the same as fig. 11.

Hereinafter, a common feature in the embodiment of the present invention is the processing of the double frame lines described in item numbers 7 and 8 in the flowchart of fig. 15. In the third embodiment, the following mode is adopted: when it is determined that the data after the start of the tact is stored in the tact start determination of item No. 5 (S35), the trigger amount holding data is stored together, and the data of the remaining trigger length is added to the file every time the remaining trigger length is acquired from the file. In this way, the processing load of the I/O, which is a problem of the conventional example, can be suppressed by storing most of the data successively instead of storing all of the data at once.

According to this configuration, the effect of suppressing the I/O processing load during the storage of the tact data and suppressing the influence of the sensor data on the real-time processing can be obtained. In the third embodiment, the processing unit 40 executes the tact start determination unit 42 in item numbers 7 to 8 (S37 to S38), as an example.

< Fourth embodiment >, a third embodiment

Fig. 16 is a flowchart showing an example of a flow of processing of the takt time division data collection system of the fourth embodiment, in which the takt time division data collection system of the fourth embodiment is similar to the configuration shown in fig. 7, and the power supply configuration and the sensor installation site of the target device are similar to the examples described with reference to fig. 8. In addition, the block diagram of the process of the tact divided data collection system of the fourth embodiment is the same as fig. 11.

Hereinafter, a common feature in the embodiment of the present invention is the processing of the double frame lines described in item numbers 1 and 2 in the flowchart of fig. 16. In the fourth embodiment, the determination of the tact reference waveform and the trigger waveform is not performed by the analysis performed by the manual work of the system builder, but is automated by the system. This can suppress the man-hour of system construction work. Hereinafter, the respective activation methods will be described.

In the automation of the tact reference waveform determination described in item No. 1, continuous data including a plurality of tacts is used as learning data, and the autocorrelation function shown in fig. 17 is first derived.

Here, n represents the data amount of continuous data, and x _i represents continuous data. In contrast to the autocorrelation function (ACF (t)), which is the similarity between waveforms, the ACF (t) calculates the similarity transition when one waveform is shifted from another one by one. In the case of calculating ACF (t) for continuous data containing a plurality of tacts, the value of ACF (t) has a peak value at the same period as the tact length. By applying this property, determination of the tact reference waveform can be performed.

The automation of the trigger waveform determination then uses the previous continuous data and the determined tact reference waveform. First, NCC is calculated using the continuous data and the tact reference waveforms, and the number of tact reference waveforms included in the continuous data is calculated from the similarity. Then, the temporary trigger waveform is cut out from the head of the tact reference waveform by a predetermined length. And calculating NCC through the continuous data and the temporary trigger waveforms, and calculating the number of the temporary trigger waveforms contained in the continuous data according to the similarity. Finally, whether the temporary trigger waveform can be a formal trigger waveform is determined based on whether the number of tact reference waveforms matches the number of temporary trigger waveforms. When the numbers are not identical, the temporary trigger waveform is extended from the predetermined length by a predetermined amount, and the same calculation is performed, and the process is repeated until the numbers are identical. If the calculation is performed until the temporary trigger waveform reaches the length of the tact reference waveform, the relationship that the number of temporary trigger waveforms gradually approaches the number of tact reference waveforms is obtained from a state in which the number of temporary trigger waveforms is excessive, and the initial temporary trigger waveform length consistent with the number of tact reference waveforms becomes the length of the trigger waveform to be determined, thereby realizing automatic determination of the trigger waveform.

According to the present configuration in which the above two automation methods are carried out, the tact reference waveform determination and the trigger waveform determination, which must be performed by manual work, can be automated, and the man-hour of manual work can be suppressed. In item numbers 1 to 2 (S41 to S42), the processing unit 40 executes the tact start determination unit 42.

< Fifth embodiment >, a third embodiment

Fig. 18 is an example of a current waveform and a waveform of a statistic thereof to be subjected to the takt-time division data collection system according to the fifth embodiment, fig. 19 is a flowchart showing an example of a flow of processing of the takt-time division data collection system according to the fifth embodiment, the takt-time division data collection system according to the fifth embodiment is the same as the configuration shown in fig. 7, and a power supply configuration and a sensor installation site of the apparatus to be subjected to the processing are the same as the example described using fig. 8. In addition, the block diagram of the process of the tact divided data collection system of the fifth embodiment is the same as fig. 11.

Hereinafter, a common feature in the embodiment of the present invention is the processing of the double frame line described by item number 1 in the flowchart of fig. 19. In the embodiment described above, the simplified sensor data is processed, but the current value actually flowing in the motor is an ac waveform as shown in fig. 18A, and the magnitude of the value changes in a short period. Therefore, if the degree of similarity such as NCC is calculated from the original waveform, the accuracy may be lowered. Therefore, in the fifth embodiment, a step of simplifying the ac waveform shown in fig. 18A by using statistics as shown in item number 1 (S51) is added. In the fifth embodiment, simplification is made from the ac waveform with moving RMS, as an example. The moving RMS is a method of calculating RMS shown in equation 1 (fig. 6) for each predetermined section, and a waveform shown in fig. 18B is obtained. In the present embodiment, the waveform of fig. 18B is used for calculation such as the production tact start judgment, and the ac waveform of fig. 18A is stored in the final tact file storage.

According to this configuration, even if the sensor data is a complex waveform such as an ac waveform, the accuracy of the takt time division can be maintained. In item No. 1 (S51), the processing unit 40 executes the tact start determination unit 42.

< Sixth embodiment >

Fig. 20 is a flowchart showing an example of a flow of processing of the takt time division data collection system of the sixth embodiment, the takt time division data collection system of the sixth embodiment being similar to the configuration shown in fig. 7, and the power supply configuration of the target device being similar to the example described using fig. 8. In addition, the block diagram of the process of the tact divided data collection system of the sixth embodiment is the same as fig. 11. In the sixth embodiment, a case where a plurality of acquired sensor data are used is targeted, for example, a case where all of the sensor data of the first to sixth motors in fig. 8 are acquired.

Hereinafter, a common feature in the embodiment of the present invention is the processing of the double frame line described by item number 1 in the flowchart of fig. 20. In the embodiment described above, the sensor data is assumed to be a single channel, but in the present embodiment, the data of a plurality of channel amounts is acquired in a lump. In the present embodiment, instead of performing the tact start determination calculation for each channel as described in item No. 1 (S71), arbitrary 1-channel data is selected to perform calculation such as tact start determination, and the result is used to perform tact division on the data of all channels.

According to this configuration, even when the number of channels increases, the calculated amount of the takt time division data collection system can be suppressed to the same extent as 1 channel. In item No.1 (S71), the processing unit 40 executes the tact start determination unit 42.

< Seventh embodiment >, a third embodiment

Fig. 21 is a diagram showing an example of an abnormality detection system and an object thereof according to the seventh embodiment, and fig. 22 and 23 are a flowchart and a block diagram showing an example of a processing flow of the abnormality detection system according to the seventh embodiment. The power supply structure and the sensor installation site of the target device are the same as those of the example described with reference to fig. 8.

The seventh embodiment is characterized by the double frame line processing described in item 10 in the flowchart of fig. 22. Other processes may be performed in the same manner as described above, and in the present embodiment, the abnormality detection function of the implementation device is provided as shown in item number 10 (S100) using the data after the takt time division described in the previous embodiment.

Specifically, the processing unit 40 executes an abnormality detection algorithm to determine whether or not the stored data is normal. Then, the processing unit 40 outputs the result to the detection result display unit 50. Here, the detection result display unit 50 can be configured using an appropriate display. The processing unit 40 can execute an appropriate program in the display processing related to the abnormality detection. The display mode is not particularly limited as long as the operator or the like can grasp the result of abnormality detection.

According to the present embodiment, there is provided an abnormality detection system including: a tact-dividing data collection module that collects data of each tact; and an abnormality detection module that makes a normal/abnormal determination using the obtained data for each tact. According to this configuration, the effect is obtained that the divided tact data can be effectively utilized, and abnormality detection can be performed while data is collected. In item number 10 (S100), the processing unit 40 executes the abnormality detection unit 43.

As described above, according to the embodiments of the present invention, the following problems can be solved: in the abnormality detection of the additional sensor for the discrete system steps such as machining in units of workpieces and robot handling, it is necessary to perform takt cutting after data recording, but in the case where it is difficult to acquire a trigger signal, takt cutting is performed by pattern matching using the entire takt reference waveform, and therefore, from the viewpoint of processing load, it is necessary to prepare a processing device separately from the recording apparatus, and the cost increases.

The embodiments of the present invention have been described in detail above, but the present invention is not limited to the above embodiments, and various design changes can be made without departing from the spirit of the present invention described in the scope of the patent claims. For example, the above-described embodiments are described in detail for the purpose of easily understanding the present invention, and are not limited to the configuration in which all the components described are necessarily provided. In addition, a part of the structure of one embodiment may be replaced with the structure of another embodiment, and the structure of another embodiment may be added to the structure of one embodiment. Further, with respect to a part of the structure of each embodiment, addition, deletion, and substitution of other structures can be performed.

As an example, the tact divided data collection system 3 and the abnormality detection system 5 are configured using 1 or more computers, and as shown in fig. 24, a system including a processor, a storage device, and an interface for inputting and outputting data can be used. The processing unit 40 described in the embodiment may be configured using a processor, and the storage unit 30 may be configured using a storage device. Here, as an example of the processor, a CPU (Central Processing Unit ) is given, but other semiconductor devices may be used. As an example of the storage device, SSD (Solid STATE DRIVE) is given. The storage device may include a RAM (Random Access Memory: random access memory) for reading a program by the processor, or an HDD (HARD DISK DRIVE: hard disk drive) or the like may be used as the storage device.

The sensor data acquisition unit 41, the tact start determination unit 42, and the abnormality detection unit 43 described in the embodiment are implemented by, for example, a program executed by a processor. The tact reference waveform storage unit 31, the trigger waveform storage unit 32, and the tact data storage unit 33 are arranged as databases in the storage device. Further, other data may be stored in the storage device, and for example, data related to abnormality detection may be stored.

As an example, the tact-time division data collection system 3 and the abnormality detection system 5 are disposed at a site where the robot performs work, but may be disposed on a cloud different from the site. Thereby, data can be managed on the cloud.

The object device that collects the tact data is not limited to the configuration of the robot described in the embodiment. For example, the present invention may be applied to various industrial equipment that performs various operations in a repetitive operation.

The tact-divided-data collection system 3 and the abnormality detection system 5 may acquire data from other types of sensors that output the same value in each tact, and process the data. For example, data based on a force sensor may be acquired and processed in the same manner.

The abnormality detection system 5 may perform other notification when notifying an abnormality to an operator or the like. For example, notification by a lamp or a buzzer may be performed.

The tact-time division data collection system 3 and the abnormality detection system 5 may input and output data by using a connected removable nonvolatile memory device.

Symbol description

3: Data collection system is cut apart to takt

30: Storage unit

31: Takt reference waveform storage unit

32: Trigger waveform storage unit

33: Tact data storage unit

40: Processing unit

41: Sensor data acquisition unit

42: Tact start determination unit

400: Distribution board

500: Control box

510: Power supply circuit

520: Control substrate

600: Robot part

601: First operating shaft

602: Second operating shaft

603: Third operating shaft

604: Fourth operating shaft

605: Fifth operation shaft

606: And a sixth operating shaft.

Claims (12)

1. A system for collecting data of takt time division is characterized in that,

The tact divided data collection system is provided with a processor,

The processor performs the following processing:

Determining whether or not a trigger waveform is included in the takt waveform in the time-series data acquired from the sensor, based on the trigger waveform stored in advance in the storage device as a part of the takt waveform of the repeating waveform;

When it is determined that the takt waveform includes a trigger waveform, data of the takt waveform including the trigger waveform is collected.

2. The tact split data collection system of claim 1, wherein,

The storage means stores a waveform of a fixed amount of time from the beginning of the tact waveform as a trigger waveform.

3. The tact split data collection system of claim 1, wherein,

The storage means stores a waveform of a fixed amount of time in a middle portion of the tact waveform as a trigger waveform,

The processor collects data of the tact waveform before the start of the trigger waveform based on the temporarily stored time-series data.

4. The tact split data collection system of claim 1, wherein,

The storage means stores a waveform of a fixed amount of time of a terminal portion of the tact waveform as a trigger waveform,

The processor collects data of the tact waveform before the start of the trigger waveform based on the temporarily stored time-series data.

5. The tact split data collection system of claim 1, wherein,

The processor performs the following processing:

In the determination, data of a trigger waveform length amount is taken out from the tact waveform, and a similarity with the stored trigger waveform is calculated; and

When the value based on the similarity is determined to be similar, the data of the remaining length is written to the file after the data of the fetched length is written to the file.

6. The tact split data collection system of claim 5, wherein,

The similarity is calculated based on the cross-correlation.

7. The tact split data collection system of claim 1, wherein,

The storage device stores trigger waveforms as follows: in the continuous data including a plurality of takt waveforms, the number of takt waveforms is calculated, and the trigger waveforms determined by length are adjusted so that the number of trigger waveforms cut from the takt waveforms included in the continuous data matches the calculated number of takt waveforms.

8. The tact split data collection system of claim 7, wherein,

An autocorrelation function is used to determine the beat waveform in the continuous data.

9. The tact split data collection system of claim 1, wherein,

The processor calculates statistics of time series data taken from the sensors and processes using the calculated statistics.

10. The tact split data collection system of claim 1, wherein,

In the case where each time-series data is acquired from a plurality of sensors, the processor performs processing using the time-series data acquired from any one of the sensors.

11. The tact split data collection system of claim 1, wherein,

The takt time division data collection system is configured on the cloud.

12. An abnormality detection system, characterized in that,

The abnormality detection system includes a tact divided data collection module and an abnormality detection module,

The takt time division data collection module includes:

a sensor data acquisition unit that acquires time-series data from a sensor;

A tact waveform storage section that stores one of the repeated waveforms included in the learning-time data as a tact reference waveform;

A trigger waveform storage section that stores a fixed amount of time waveform from the beginning of the tact reference waveform as a trigger waveform that is a tact start trigger;

A tact start determination unit that includes a process of extracting data of a trigger waveform length amount from data at the time of operation and calculating a similarity with the trigger waveform; and

A tact data storage unit for storing the operation time data as a file for each tact length based on the calculation result,

The abnormality detection module makes a normal/abnormal determination using the obtained data for each tact.