JP2023113056A - Tact-divided data collecting system and abnormality detecting system - Google Patents

Abstract

To provide a tact-divided data collecting system and an abnormality detecting system capable of realizing tact extraction at the same time as logging and reducing the number of devices, thereby collecting the tact-divided data at a low cost.SOLUTION: The tact-divided data collecting system for solving the above problem is provided with a processor. The processor of the tact-divided data collecting system, on the basis of a trigger waveform which is a part of a tact waveform being a repetitive waveform, and which is previously stored in a storage device, performs the processing of determining whether the tact waveform in time series data acquired from a sensor includes the trigger waveform. Then, when the processor of the tact-divided data collecting system determines that the tact waveform includes the trigger waveform, the tact waveform data including the trigger waveform is collected.SELECTED DRAWING: Figure 6

Description

The present invention relates to a takt-divided data collection system and an anomaly detection system.

The failure frequency of industrial equipment varies greatly depending on the number of years since its introduction. 1-2 years after introduction, the initial failure period when failures occur due to defects in design or manufacturing; A wear-out failure period, in which failures increase due to the life of the machine, is statistically shown. The initial failure period is supported by the manufacturer's response to initial defects, but the wear-out failure period depends on the usage conditions of the machine and the user's maintenance frequency, making it difficult to predict when it will arrive. If this is not grasped, there is a problem that the operating rate will be lowered, and productivity will likely decline. With regard to equipment during this period, it is often difficult to introduce the latest monitoring solutions from manufacturers because the models are outdated, so there is a need for a solution that adds a new general-purpose sensor to monitor and detect anomalies. tend to increase. Such add-on sensors include a temperature sensor, a vibration sensor, and a current sensor for measuring a motor drive current in the case of a device driven by a motor. If the target equipment is a processing machine for each work such as machining, or a vertical articulated robot that transports a work, the measurement data will be a repetition of "takt" that indicates a series of processes, so the repeated takt is used for abnormality detection. It is common to use means for detecting the difference by comparing them.

Patent Literature 1 proposes a method of realizing takt division by performing pattern matching between a predetermined takt reference waveform and acquired data.

JP 2020-187502 A

In the case of abnormality detection by an add-on sensor, it is often difficult to obtain a trigger signal for starting takt from the control system due to the desire to minimize the impact on existing equipment. Since there are many variations between takts due to work waiting and work waiting for peripheral devices, takt division may be missed in fixed cycle extraction with takt length. The issue is whether it can be achieved with high accuracy.

In order to meet the above requirements, the system described in Patent Document 1 provides a learning phase before operation, identifies and stores the tact reference waveform from data acquired in advance, and Pattern matching is performed between the acquired data and the takt reference waveform, and if the degree of matching exceeds a certain amount, it is determined that there is a takt, and the data is cut out as takt data. However, this approach has two drawbacks. The first is that pattern matching is performed on the entire takt reference waveform, so the computational load is high and it is not easy to implement it simultaneously in data logging equipment. Secondly, when matching is established, the extracted data is handled all at once, so if the system tries to save the data in a file, the data I/O load is large. is likely to affect the data logging equipment, so it may not be easy to implement simultaneously in the data logging equipment. Due to the above two drawbacks, it is necessary to prepare a device for dividing the takt time separately from the data logging device, which raises the problem of an increase in device cost.

Therefore, the present invention has been made in view of the above points, and its purpose is to realize takt extraction at the same time as logging, reduce the number of devices, and collect takt divided data at low cost. That is.

According to a first aspect of the present invention, the following takt-divided data collection system is provided. That is, the takt-divided data collection system includes a processor. The processor determines whether the trigger waveform is included in the tact waveform in the time-series data acquired from the sensor, based on the trigger waveform that is part of the tact waveform that is a repetitive waveform and is stored in advance in the storage device. When the processor determines that the tact waveform includes the trigger waveform, it collects data of the tact waveform including the trigger waveform.

According to a second aspect of the present invention, the following anomaly detection system is provided. That is, the anomaly detection system includes a takt-divided data collection module and an anomaly detection module. The tact split data collection module includes a sensor data acquisition unit, a tact waveform storage unit, a trigger waveform storage unit, a tact start determination unit, and a tact data storage unit. The sensor data acquisition unit acquires time series data from the sensor. The takt waveform storage unit stores one of the repetitive waveforms included in the learning data as a takt reference waveform. The trigger waveform storage unit stores a waveform for a certain period of time from the beginning of the tact reference waveform as a trigger waveform that is a tact start trigger. The takt start determination unit includes a process of extracting data corresponding to the length of the trigger waveform from the operating data and calculating the degree of similarity with the trigger waveform. The takt data storage unit stores operation data as a file for each takt length according to the calculation result. The anomaly detection module performs normality/abnormality determination using the obtained data for each takt.

According to the present invention, there is provided a system capable of realizing takt extraction at the same time as logging and collecting takt divided data at low cost with a reduced number of devices. Problems, configurations, and effects other than those described above will be clarified by the following description of the mode for carrying out the invention.

FIG. 10 is a diagram showing the external appearance of a data collection system and a target device according to a conventional example; FIG. 10 is a diagram showing a power circuit of a device targeted by a data collection system according to a conventional example; FIG. 10 is a diagram showing current waveforms obtained by sensing a target device by the data collection system according to the conventional example and the present embodiment, and is a diagram showing an example of current waveforms including multiple takts. FIG. 10 is a diagram showing current waveforms obtained by sensing target devices by the data collection system according to the conventional example and the present embodiment, and is a diagram showing an example of a reference waveform indicating one takt. FIG. 5 is a flow chart showing the flow of processing in a conventional takt divided data collection system. FIG. 11 is a block diagram of a takt divided data collection system according to a conventional example, and is a block diagram during learning; FIG. 11 is a block diagram of a takt divided data collection system according to a conventional example, and is a block diagram during operation. 1 is a diagram showing the appearance of an example of a takt-divided data collection system and a target device according to an embodiment; FIG. FIG. 2 is a diagram showing an example of a power circuit of a device targeted by the data collection system according to the embodiment; FIG. 4 is a diagram showing an example of a trigger waveform indicating the start of acquisition of one takt of a current waveform obtained by sensing a target device by the takt-divided data collection system according to the first embodiment; FIG. 4 is a flowchart diagram showing an example of the flow of processing of the takt divided data collection system according to the first embodiment; It is a block diagram which shows an example of a process of the takt division data collection system which concerns on embodiment, and shows a block diagram during learning. It is a block diagram which shows an example of a process of the takt division data collection system which concerns on embodiment, and shows a block diagram in operation. FIG. 4 is a diagram showing an example of a trigger waveform indicating the start of acquisition of one takt of a current waveform obtained by sensing a target device by the takt-divided data collection system according to the first embodiment; The flowchart figure which shows an example of the flow of a process of the takt division data collection system which concerns on 2nd Embodiment. The flowchart figure which shows an example of the flow of a process of the takt division data collection system which concerns on 3rd Embodiment. The flowchart figure which shows an example of the flow of a process of the takt division data collection system which concerns on 4th Embodiment. FIG. 20 is a diagram showing a current waveform obtained by sensing a target device by the takt-divided data collection system according to the fifth embodiment, and is a diagram showing an example of a raw current waveform; FIG. 20 is a diagram showing a current waveform obtained by sensing a target device by the takt-divided data collection system according to the fifth embodiment, and is a diagram showing an example of a movement RMS waveform; The flowchart figure which shows an example of the flow of a process of the takt division data collection system which concerns on 5th Embodiment. The flowchart figure which shows an example of the flow of a process of the takt division data collection system which concerns on 6th Embodiment. The figure which shows the external appearance of an example of the abnormality detection system which concerns on 7th Embodiment, and a target apparatus. The flowchart figure which shows an example of the flow of a process of the abnormality detection system which concerns on 7th Embodiment. FIG. 20 is a block diagram showing an example of processing of the anomaly detection system according to the seventh embodiment, and is a block diagram during learning; FIG. 20 is a block diagram showing an example of processing of the anomaly detection system according to the seventh embodiment, and is a block diagram during operation; FIG. 2 is a hardware configuration diagram regarding an example of a system configuration;

As described above, in the takt-divided data collection system using the mechanism of performing pattern matching with the reference waveform described in Patent Document 1, pattern matching is performed with the acquired data for the tact length and the takt reference waveform, and when the matching is established, the data It can be a format to save all at once. With such a method, as described above, the calculation load and I/O load are high, and simultaneous implementation in the logging device is not possible, which may lead to an increase in cost due to the addition of devices.

First, an example of devices and sensor data targeted by a conventional takt-divided data collection system and an example of a system configuration will be described.

FIG. 1 is a diagram showing an example of a conventional takt-divided data collection system and a target device, and the target device is a robot. In this system, the robot includes a control box 200 and a robot section 300 . A plurality of operating axes exist in the robot unit 300, and a first operating axis 301, a second operating axis 302, a third operating axis 303, a fourth operating axis 304, a fifth operating axis 305, and a sixth operating axis 306 are , can rotate in the directions of the arrows.

The takt-divided data collection system 1 is connected to the control box 200 of the target device. The takt-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 that stores a tact reference waveform and a tact data storage unit 13 that stores divided data. The processing unit 20 includes a sensor data acquisition unit 21 and a takt start determination unit 22 .

FIG. 2 is a diagram illustrating an example of a sensing target circuit of a target device of the takt-divided data collection system. Electric power is supplied from the control box 200 to each motor that is a driving motor in the robot section 300 . The robot unit 300 includes a plurality of drive motors, and the rotation of each drive motor is transmitted to a speed reducer to operate a working shaft provided ahead. A speed reducer is a mechanism that reduces the input rotational speed and increases torque instead.

In the example of FIG. 2, the control box 200 includes a power supply circuit 210 and a control board 220, and the power supplied from the switchboard 100 is delivered to the power supply circuit 210, passes through the control board 220, and reaches each motor. supplied. The control board 220 can supply power to multiple power consuming devices. As 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 section 300. .

Each drive motor includes a first reduction gear 311, a second reduction gear 312, a third reduction gear 313, a fourth reduction gear 314, a fifth reduction gear 315, a sixth reduction gear 316, are connected to each other. Each speed reducer includes a first operating shaft 301, a second operating shaft 302, a third operating shaft 303, a fourth operating shaft 304, a fifth operating shaft 305, and a sixth operating shaft 306. , are connected respectively. In the robot unit 300, the motors connected to each operating axis are driven according to the magnitude and frequency of the current output from the control board 220 according to the operation program. Therefore, it is known that these current waveforms are important physical quantities that reflect the state of deterioration of the motors of each operating shaft of the robot and the reduction gears and operating shafts that are mechanical parts connected to the motors. In the conventional example, sensing is performed by clamping a current sensor to the power supply circuit connected to the first motor.

FIG. 3 is a waveform simply showing a current waveform acquired by a current sensor, FIG. 3A is a diagram showing an example of a current waveform including multiple tacts, and FIG. 3B is a diagram showing an example of a reference waveform showing one tact. . One takt represents the minimum unit of a series of repetitive motions of the robot, and in FIG. It is a system that outputs data for one takt of 3B to a file. FIG. 4 is a diagram showing a processing flow of a conventional takt-divided data collection system, and FIG. 5 is a block diagram of the conventional takt-divided data collection system.

A conventional takt-divided data collection system will first be described from the learning process with reference to FIGS. As described above, a current sensor is clamped on the wiring between the first motor that operates the first operating shaft 301 of the robot unit 300 in FIGS. A waveform is acquired. This flow is the procedure indicated by item numbers 0 and 1 (S1 and S2) in FIG. 4, and corresponds to the block diagram during learning in FIG. 5A. The takt-divided data acquisition system 1 observes this continuous current waveform, identifies repeated takt waveforms, and stores them as takt reference waveforms.

The sensor data acquisition unit 21 collects sensor data for a predetermined time period of 100 msec to 1 sec and saves it as a file. and the takt length is assumed to be 60 seconds, for example.

In the identification step in the takt reference waveform storage unit 11, the stored 1 sec file is analyzed by linking, for example, 1 hour, that is, 3600 files, and the presence of 60 sec takt is clarified. Work to save the waveform is performed. Next, the operation process will be explained. During operation, as shown in Item No. 2 (S3) in FIG. 4, 60 temp files are acquired, combined, and held to prepare for the comparison process with the reference waveforms of Item Nos. 3 and 4 (S4 and S5). do. In the conventional example, a statistic represented by the following formula is used as a comparison means.

Here, RMS is the Root Mean Square, n is the amount of data for the tact length, and _xi is the sensor value, that is, the current value, and one value is calculated for the entire tact waveform. In Item No. 3, the RMS of the takt reference waveform identified and held in advance and the RMS of the acquired combined data for 60 seconds obtained during operation are calculated. Item No. 4 determines whether or not the RMS difference absolute value is equal to or less than a predetermined value. ], the absolute value of the RMS difference is 0.5 [Arms].

Assuming that the threshold is 0.3 [Arms], the absolute value of the RMS difference is greater than or equal to the threshold, so the process proceeds to Item No. 5 (S6). In Item No. 5, since it is assumed that the acquired combined data does not match the takt reference waveform, the oldest file is deleted in order to update the acquired combined data. By doing so, when returning to Item No. 2, one new file is added, and the takt start determination step is performed again.

This loop of item numbers 2 to 5 is repeated, and when the absolute value of the RMS difference is equal to or less than the threshold, it is assumed that the takt exists in the acquired combined data, so the process moves to item number 6 (S7). . Item No. 6 is a step of saving the acquired combined data as a tact file, and after storing it in the tact data storage unit, item No. 7 (S8) of deleting all temp files related to the acquired combined data is performed. Basically, the takt does not overlap with the next takt, so the data is reset once in this way and the next takt is waited for. In item number 0 (S1), the processing unit 20 executes the sensor data acquisition unit 21, and in item numbers 1 to 7 (S2 to S8), the processing unit 20 executes the takt start determination unit 22.

Although it is possible to build a takt-divided data collection system even in the conventional example with the above flow, in item No. 3, in order to perform statistical value calculation processing for the takt time, sensor data acquisition work that is being processed in real time etc., may be affected. Therefore, as indicated by the dotted line arrow in FIG. 4, there is a demerit that the equipment cost increases by dividing the equipment that performs the processing. In addition, since it is necessary to collectively save the data for takts even when saving the data held for takts in Item No. 6, there is a possibility that it will affect the real-time sensor data acquisition process from the viewpoint of I/O. However, there is a demerit that equipment costs may increase.

Next, an embodiment for solving such problems will be described. It should be noted that the embodiments are examples for explaining the present invention, and are appropriately omitted and simplified for clarity of explanation. The present invention can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. The position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc., in order to facilitate understanding of the invention. As such, the present invention is not necessarily limited to the locations, sizes, shapes, extents, etc., disclosed in the drawings.

All of the following embodiments are methods that can suppress the calculation load and the I/O load, and have the effect of realizing a low-cost takt-divided data collection system. From the viewpoint of cost reduction, resource saving can also be achieved.

In the following embodiments, the same reference numerals indicate the same configuration even if the figure numbers are different, and have the same action and effect. Also, the description of the content that has already been explained or similar content may be omitted.

<First Embodiment>
The takt divided data collection system in the first embodiment has the configuration shown in FIG. It can be the same as the conventional example described above. That is, the takt-divided data collection system 3 includes a storage unit 30 and a processing unit 40 . Also, the target device can be a robot, and the robot includes a control box 500 and a robot section 600 . The control box 500 includes a power supply circuit 510 and a control board 520, and power supplied from the switchboard 400 is transferred to the power supply circuit 510, passes through the control board 520, and is supplied to each motor. The robot unit 600 includes motors (621-626), reduction gears (611-616), and operating shafts (601-606). FIG. 8 is a diagram showing an example of a trigger waveform according to the first embodiment, FIG. 9 is a flowchart showing an example of the processing flow of the takt divided data collection system according to the first embodiment, and FIG. It is a block diagram showing an example of processing of the takt divided data collection system according to one embodiment.

Hereinafter, the common features of the embodiments of the present invention are the processing of double border lines and oblique lines described in item numbers 2, 4, and 7 (S12, S14, and S17) in the flow chart of FIG. . Here, during learning, the trigger waveform is identified after the tact reference waveform is identified.

This trigger waveform is data obtained by cutting out a part of the tact reference waveform, and its role is to determine whether or not there is a tact based on the amount of data equal to or less than the tact length. As shown in FIG. 10A, the tact reference waveform is identified by the same procedure as in the conventional example, and the takt reference waveform is stored in the tact reference waveform storage unit 31 .

Then, using the data, the trigger waveform storage unit 32 identifies a unique waveform as shown in FIG. 8 that exists within the takt and does not repeatedly exist within the takt. Assume that this unique waveform is, for example, the first 10 seconds. After identification, the unique waveform is stored in the trigger waveform storage section 32 as a trigger waveform. This procedure corresponds to steps 0 to 2 (S10 to S12) in FIG.

Then, during operation, as shown in item number 3 (S13) in FIG. 9, only 10 sec worth of trigger data is acquired, combined, and held. In the next item No. 4 (S14), the same statistical amount calculation and comparison as in the conventional example are performed with the data corresponding to the length of the trigger waveform. If the statistic difference absolute value is equal to or less than the threshold in Item No. 5 (S15), the process proceeds to Item No. 7 (S17). Item number 7 (S17) adds the acquired data to the data for the trigger waveform, and successively acquires the remaining data for the takt length from the file. Then, when the takt length is reached, as described in Item No. 8 (S18), collective storage as a takt file is performed. Also, all the temp files combined in Item No. 9 (S19) are deleted.

In this embodiment, as shown in FIG. 10B, the tact start determination calculation can be performed with data for the trigger waveform length shorter than the takt length. For example, as described above, when the tact length is 60 sec and the trigger waveform length is 10 sec, the tact start determination calculation can be reduced to 1/6 compared to the conventional example, and the tact division data collection system The effect of suppressing the calculation cost of 3 is obtained.

Further, as shown in FIG. 11, it is also possible to take the middle part of the tact reference waveform as the trigger waveform. need to keep Since only the trigger waveform length data is used in the takt start determination calculation, the same calculation cost reduction effect can be obtained, but considering the memory cost for temporary storage, a unique waveform from the beginning is used as the trigger waveform , it is possible to suppress the increase in memory cost. Also, based on the same point of view, the trigger waveform at the end portion of the takt reference waveform may be used.

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

<Second embodiment>
FIG. 12 is a flowchart showing the flow of processing of the takt-divided data collection system according to the second embodiment. The configuration of the power source of the device and the locations where the sensors are installed are the same as in the example described with reference to FIG. Also, the block diagram of the processing of the takt divided data collection system according to the second embodiment is the same as that of FIG.

Hereinafter, the common feature in the embodiments of the present invention is the use of normalized cross-correlation (NCC) for the temporary storage format of data described in item number 0 in the flowchart of FIG. That is.

In the first embodiment, the data is temporarily saved in a file, while in the second embodiment, the data is temporarily saved in a memory. By doing so, it is possible to suppress the I/O load when saving the obtained data. However, since an extra memory storage area is required, it is necessary to select each configuration according to the device to be used. The main effect of the second embodiment is achieved by using NCC, which is another feature. Here, NCC is calculated by the following formula.

Here, n is the amount of data for the length of the trigger waveform, _{xi is} the acquired data for the length of the trigger waveform, and _yi is the previously identified trigger waveform. NCC is a metric used to evaluate the similarity between two waveforms. By using the NCC, compared with the case of using the statistics shown in the first embodiment, the similarity of the entire waveform can be evaluated, and the threshold can be determined by a relative value of the similarity. be. For example, in NCC, the degree of similarity between two waveforms can be calculated as a numerical value between 0 and 1, so the effect of reducing the number of SI man-hours for selecting a threshold for each object can be obtained.

According to this configuration, it is possible to reduce the number of man-hours for setting a threshold value for tact start determination while improving the accuracy of tact start determination. In addition, although the determination method using NCC has been described in this embodiment, it is also possible to use other typical similarity calculation methods between waveforms, such as the sum of squared differences (SSD) and the sum of absolute differences (SAD). It is possible to improve the accuracy of the takt start determination as in the case of using the NCC.

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

<Third Embodiment>
FIG. 13 is a flowchart showing an example of the flow of processing of the takt-divided data collection system according to the third embodiment. The power source configuration and sensor installation locations of the device are the same as the example described with reference to FIG. 7 . Also, the block diagram of the processing of the takt divided data collection system according to the third embodiment is the same as that of FIG.

Hereinafter, the common feature of the embodiments of the present invention is the processing of the double frame line and oblique line filling described in item numbers 7 and 8 in the flow chart of FIG. 13 . In the third embodiment, when saving data after it is determined that the tact has started in the tact start determination in item number 5 (S35), the data held for the trigger is collectively saved, and the data for the rest of the trigger length after that is stored from the file. It is a method of sequentially appending to the file each time it is acquired. By saving most of the data sequentially instead of saving all the data collectively in this way, it is possible to suppress the I/O processing load, which has been a problem in the conventional example.

According to this configuration, it is possible to suppress the I/O processing load during storage of takt data, thereby suppressing the influence of sensor data acquisition on real-time processing. In the third embodiment, as an example, the processing unit 40 executes the takt start determination unit 42 in item numbers 7 and 8 (S37 and S38).

<Fourth Embodiment>
FIG. 14 is a flowchart showing an example of the processing flow of the takt-divided data collection system according to the fourth embodiment. The power supply configuration and sensor installation locations of the device are the same as the example described with reference to FIG. 7 . Also, the block diagram of the processing of the takt-divided data collection system according to the fourth embodiment is the same as that of FIG.

A common feature of the embodiments of the present invention is the processing of double border lines and oblique line filling described in item numbers 1 and 2 in the flow chart of FIG. In the fourth embodiment, identification of the tact reference waveform and the trigger waveform is automated by the system rather than manual analysis by the system builder. By doing so, it is possible to reduce the man-hours required for system construction work. Each automation method will be described below.

In the automation of takt reference waveform identification described in Item No. 1, an autocorrelation function represented by Equation (3) is first derived using continuous data containing a plurality of takts as learning data.

Here, n is the data amount of the continuous data, and _xi represents the continuous data. Regarding the autocorrelation function (ACF(t)), cross-correlation is the degree of similarity between waveforms, while ACF(t) calculates the similarity transition when one of the waveforms is slightly shifted. . When ACF(t) is calculated for continuous data including multiple takts, the value of ACF(t) has a peak at the same period as the takt length. By applying this property, it is possible to identify the takt reference waveform.

Subsequently, automatic trigger waveform identification uses the continuous data and the identified takt reference waveform. First, the NCC is calculated from the continuous data and the tact reference waveform, and the number of tact reference waveforms included in the continuous data is calculated from the degree of similarity. Subsequently, a provisional trigger waveform is obtained by cutting out a specified length from the head of the takt reference waveform. NCC is calculated from the continuous data and provisional trigger waveforms described above, and the number of provisional trigger waveforms included in the continuous data is calculated from the degree of similarity. Finally, it is determined whether or not the provisional trigger waveform can be a formal trigger waveform by determining whether the number of tact reference waveforms and the provisional trigger waveform number match. If the numbers do not match, the provisional trigger waveform is extended from the specified length by a specified amount, and similar calculations are performed until the numbers match. If this calculation is performed until the provisional trigger waveform reaches the tact reference waveform length, a relationship is obtained in which the number of provisional trigger waveforms is excessive and gradually approaches the tact reference waveform number. The first matching tentative trigger waveform length becomes the length of the trigger waveform to be identified, and automatic identification of the trigger waveform is achieved.

According to this configuration in which both of the above automation methods are installed, the effect of automating the takt reference waveform identification and trigger waveform identification, which had to be performed manually, can be obtained, thereby reducing the number of manual steps. In item numbers 1 and 2 (S41 and S42), the processing unit 40 executes the takt start determination unit .

<Fifth Embodiment>
FIG. 15 shows an example of a current waveform and its statistic waveform targeted by the takt-divided data collection system according to the fifth embodiment, and FIG. 7 is a flowchart showing an example, where the takt divided data collection system in the fifth embodiment has the same configuration as that shown in FIG. 6, and the power supply configuration and sensor installation locations of the target device are one example explained using FIG. is similar to Also, the block diagram of the processing of the takt divided data collection system according to the fifth embodiment is the same as that of FIG.

Below, the common feature of the embodiments of the present invention is the processing of double border lines and oblique line filling described in Item No. 1 in the flow chart of FIG. 16 . Although the embodiments so far have dealt with simplified sensor data, the current value actually flowing through the motor is an AC waveform as shown in FIG. 15A, and the magnitude of the value changes in a short cycle. Therefore, if a similarity like NCC is calculated with the waveform as it is, the precision may be lowered. Therefore, in the fifth embodiment, a step of simplifying the AC waveform as shown in FIG. 15A using statistics is added as shown in Item No. 1 (S51). In the fifth embodiment, as an example, the AC waveform is simplified by moving RMS. Moving RMS is a method of calculating the RMS represented by Equation 1 for each predetermined section, and a waveform such as that shown in FIG. 15B is obtained. In the present embodiment, the waveform in FIG. 15B is used in calculations such as takt start determination, and the AC waveform in FIG. 15A is saved in the final takt file saving.

According to this configuration, even if the sensor data has a complex waveform such as an AC waveform, the effect of maintaining the accuracy of the takt division can be obtained. In item number 1 (S51), the processing unit 40 executes the takt start determination unit 42. FIG.

<Sixth embodiment>
FIG. 17 is a flowchart showing an example of the flow of processing of the takt-divided data collection system according to the sixth embodiment. The power supply configuration of the device is the same as the example described with reference to FIG. Also, the block diagram of the processing of the takt divided data collection system according to the sixth embodiment is the same as that of FIG. The sixth embodiment is intended for a case in which a plurality of sensor data are acquired, for example, a case where all sensor data from the first motor to the sixth motor in FIG. 7 are acquired.

Below, the common feature of the embodiments of the present invention is the processing of double border lines and oblique line filling described in Item No. 1 in the flow chart of FIG. 17 . In the embodiments up to this point, it is assumed that the sensor data is of a single channel, but in the present embodiment, the case where the data of a plurality of channels is collectively acquired is targeted. In this embodiment, as described in Item No. 1 (S71), the takt start determination calculation is not performed for each channel, but one of the channel data is selected and the takt start determination etc. are calculated. The result is used to divide the data of all channels into takt time.

According to this configuration, even when the number of channels is increased, the effect of suppressing the amount of calculation of the takt-divided data collection system to the same extent as in the case of one channel is obtained. In item number 1 (S71), the processing unit 40 executes the takt start determination unit 42. FIG.

<Seventh embodiment>
FIG. 18 is a diagram showing an example of an anomaly detection system according to the seventh embodiment and its target, and FIGS. 19 and 20 are a flowchart and a block diagram showing an example of the processing flow of the anomaly detection system according to the seventh embodiment is. The power supply configuration and sensor installation locations of the target device are the same as the example described with reference to FIG. 7 .

The feature of the seventh embodiment is the processing of the double frame line and hatching described in item No. 10 in the flow chart of FIG. 19 . Other processing can be the same as the content described above. , and has the function of detecting anomalies in the equipment.

Specifically, the processing unit 40 executes an anomaly detection algorithm to determine whether the stored data is normal or not. 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. It should be noted that the processing unit 40 can execute a suitable program in display processing related to abnormality detection. Moreover, the mode of display is not particularly limited as long as the operator or the like can grasp the result of the abnormality detection.

According to the present embodiment, an anomaly detection system is provided that includes a takt-divided data collection module that collects data for each takt and an anomaly detection module that performs normality/abnormality determination using the obtained data for each takt. be done. According to this configuration, by utilizing the divided tact data, it is possible to obtain the effect of being able to perform data collection and abnormality detection at the same time. In item number 10 (S100), the processing unit 40 executes the abnormality detection unit 43. FIG.

As described above, according to the embodiment of the present invention, takt extraction after data logging is essential for abnormality detection by an add-on sensor for discrete processes such as machine processing for each workpiece and robot transfer. When it was difficult to acquire the trigger signal, the entire takt reference waveform was used to extract the takt by pattern matching. From the viewpoint of the processing load, it was necessary to prepare a processing device separate from the logging device, which increased costs. problem can be solved.

Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments, and various designs can be made without departing from the spirit of the invention described in the claims. Changes can be made. For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations. Also, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with another configuration.

As an example, the takt-divided data collection system 3 and the anomaly detection system 5 are configured using one or more computers, and as shown in FIG. and a system comprising: The processing unit 40 described in the embodiment can be configured using a processor, and the storage unit 30 can be configured using a storage device. Here, the processor is, for example, a CPU (Central Processing Unit), but other semiconductor devices may be used. An example of the storage device is an SSD (Solid State Drive). The storage device may include a RAM (Random Access Memory) from which the processor reads programs, or may be a HDD (Hard Disk Drive) or the like.

The sensor data acquisition unit 41, the takt start determination unit 42, and the abnormality detection unit 43 described in the embodiment are programs 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 in a storage device as a database. Note that the storage device may store other data, for example, data related to abnormality detection.

As an example, the takt-divided data collection system 3 and the anomaly detection system 5 are arranged at the site where the robot works, but they may be arranged on a cloud different from the site. This allows data to be managed on the cloud.

The target device for collecting tact data is not limited to the configuration of the robot described in the embodiment. For example, it may be other different industrial equipment that performs multiple types of operations in a repetitive operation.

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

The anomaly detection system 5 may notify an operator or the like of an anomaly in another manner. For example, a lamp or buzzer may be used for notification.

The takt-divided data collection system 3 and the abnormality detection system 5 may input/output data using a connected portable non-volatile storage device.

3: Takt division data collection system,
30: storage unit,
31: tact reference waveform storage unit;
32: trigger waveform storage unit;
33: takt data storage unit,
40: processing unit,
41: sensor data acquisition unit,
42: takt start determination unit,
400: Switchboard,
500: control box,
510: power circuit,
520: control board,
600: robot unit,
601: first working axis,
602: second operating axis,
603: third working axis,
604: fourth working axis,
605: fifth operating axis,
606: Sixth working axis.

Claims (12)

with a processor
The processor
Determining whether the trigger waveform is included in the tact waveform in the time-series data acquired from the sensor based on a trigger waveform that is part of the tact waveform that is a repetitive waveform and is stored in advance in a storage device,
When it is determined that the tact waveform includes a trigger waveform, collecting data of the tact waveform including the trigger waveform;
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The storage device
Memorize the waveform for a certain period of time from the beginning of the tact waveform as the trigger waveform.
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The storage device
A waveform for a certain period of time in the middle part of the takt waveform is stored as a trigger waveform,
The processor
Collecting tact waveform data up to the start of the trigger waveform based on the temporarily stored time-series data;
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The storage device
A waveform for a certain period of time at the end of the tact waveform is stored as a trigger waveform,
The processor
Collecting tact waveform data up to the start of the trigger waveform based on the temporarily stored time-series data;
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The processor
In the determination, extracting data for the length of the trigger waveform from the tact waveform and calculating the similarity with the stored trigger waveform,
If it is determined to be similar based on the similarity value, after writing the extracted length of data to the file, write the remaining length of data to the file.
A takt divided data collection system characterized by: The takt divided data collection system according to claim 5,
the similarity is calculated based on cross-correlation;
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The storage device
In continuous data containing multiple tact waveforms, the number of tact waveforms is calculated,
storing the identified trigger waveform whose length is adjusted so that the number of trigger waveforms extracted from the tact waveforms included in the continuous data matches the calculated number of tact waveforms;
A takt divided data collection system characterized by:
The takt divided data collection system according to claim 7,
A tact waveform in continuous data is identified using an autocorrelation function,
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The processor
Calculate the statistics of the time-series data obtained from the sensor, and perform processing using the calculated statistics.
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
When acquiring time-series data from multiple sensors,
The processor
Perform processing using time-series data acquired from any one sensor,
A takt divided data collection system characterized by: The takt divided data collection system according to claim 1,
The tact split data collection system is arranged on the cloud,
A takt divided data collection system characterized by: a sensor data acquisition unit that acquires time-series data from a sensor;
a takt waveform storage unit that stores one of the repetitive waveforms included in the learning data as a takt reference waveform;
a trigger waveform storage unit that stores a waveform for a certain period of time from the head of the tact reference waveform as a trigger waveform that is a tact start trigger;
a tact start determination unit including processing for extracting data for the length of the trigger waveform from data during operation and calculating the degree of similarity with the trigger waveform;
a takt data storage unit that saves operation data as a file for each takt length according to the calculation result;
a takt-splitting data collection module comprising
An anomaly detection module that performs normal/abnormal judgment using the obtained data for each takt,
An anomaly detection system characterized by:

Images