Condition monitoring device and condition monitoring method

Abstract

A current-related basis data acquisition unit acquires first current-related basis data and second current-related basis data pertaining to the current of a motor for driving a robot. A basis temperature acquisition unit acquires a first basis temperature and a second basis temperature that are in effect during the aforementioned acquisition of data. A current-related data acquisition unit acquires current-related data pertaining to the current of the motor. A temperature acquisition unit acquires a data acquisition temperature that is in effect during the aforementioned acquisition of data. On the basis of the first current-related basis data, the second current-related basis data, the first basis temperature, the second basis temperature, and the data acquisition temperature, a correction unit corrects the current-related data so as to correspond to a prescribed evaluation reference temperature. A robot state evaluation unit evaluates the state of the robot using the corrected current-related data.

Description

[Technical Field]

[0001] This disclosure primarily relates to a condition monitoring device that monitors the status of a robot and assists in robot maintenance. [Background technology]

[0002] When industrial robots are operated repeatedly in factories and other facilities, deterioration of various parts of the robot (e.g., mechanical components) is unavoidable. As this situation progresses, the robot will eventually fail. Since a robot failure resulting in a prolonged production line shutdown can cause significant losses, there is a strong need to perform maintenance before robot failures occur. On the other hand, performing maintenance frequently is difficult from the perspective of maintenance costs and other factors.

[0003] Patent Document 1 discloses a robot controller. In the configuration of Patent Document 1, data from the robot controller can be stored externally and analyzed by an analysis device. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2007-190663 [Overview of the project] [Problems that the invention aims to solve]

[0005] The robot data analyzed in Patent Document 1 may include data affected by temperature. However, Patent Document 1 does not describe any method for suppressing the influence of temperature changes on the data.

[0006] This disclosure is made in light of the above circumstances, and its main purpose is to effectively suppress the effects of temperature changes and to better detect signs of robot failure. [Means for solving the problem]

[0007] The problems to be solved by the present disclosure are as described above. Next, the means for solving this problem and its effects will be described.

[0008] According to a first aspect of the present disclosure, a state monitoring device having the following configuration is provided. That is, this state monitoring device monitors the state of an industrial robot capable of reproducing a predetermined operation. The state monitoring device includes a current-related basic data acquisition unit, a basic temperature acquisition unit, a current-related data acquisition unit, a temperature acquisition unit, a correction unit, and a robot state evaluation unit. The current-related basic data acquisition unit acquires first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, regarding the current of a motor that drives the industrial robot. The basic temperature acquisition unit acquires a first basic temperature, which is the temperature at the time of acquisition of the first current-related basic data, and a second basic temperature, which is the temperature at the time of acquisition of the second current-related basic data. The current-related data acquisition unit acquires current-related data, which is at least one current data or statistical data of the current data, regarding the current of the motor. The temperature acquisition unit acquires a data acquisition temperature, which is the temperature at the time of acquisition of the current-related data. The correction unit performs , so that it corresponds to the data at the evaluation standard temperature. correction on the current-related data based on the first current-related basic data, the second current-related basic data, the first basic temperature, the second basic temperature, and the data acquisition temperature. The robot state evaluation unit uses the corrected current-related data to The aforementioned industrial evaluate the Determination of whether or not there is an abnormality robot. The robot condition evaluation unit calculates the root mean square of the current, the maximum current, the minimum current, the value obtained by subtracting the current value of the lower peak from the current value of the higher peak of the current waveform, or the larger of the absolute values ​​of the maximum and minimum currents, based on the corrected current-related data, and determines whether or not there is an abnormality in the industrial robot based on the calculation results.

[0009] A second aspect of this disclosure provides the following state monitoring method. That is, this state monitoring method monitors the state of an industrial robot capable of reproducing predetermined operations. For the current of the motor driving the industrial robot, first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, are acquired. A first base temperature, which is the temperature at the time the first current-related basic data is acquired, and a second base temperature, which is the temperature at the time the second current-related basic data is acquired, are acquired. For the current of the motor, current-related data, which is at least one current data or statistical data of the current data, are acquired. A data acquisition temperature, which is the temperature at the time the current-related data is acquired, is acquired. Based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature, the current-related data is processed. , so that it corresponds to the data at the evaluation standard temperature. Perform a correction. The current-related data after correction is... The aforementioned industrial robot Determination of whether or not there is an abnormality It is used for this purpose. Based on the corrected current-related data, the mean square of the current, the maximum current, the minimum current, the value obtained by subtracting the current value of the lower peak from the current value of the higher peak of the current waveform, or the larger of the absolute values ​​of the maximum current and the absolute values ​​of the minimum current are calculated, and a determination is made as to whether or not there is an abnormality in the industrial robot based on the results of this calculation.

[0010] This allows for the evaluation of the robot's condition while suppressing the effects of temperature changes through compensation. Consequently, it enables the accurate detection of early signs of robot failure. This allows for evaluation using statistical measures while eliminating the effects of temperature changes. It enables evaluation under consistent temperature conditions. [Effects of the Invention]

[0011] According to this disclosure, the effects of temperature changes can be effectively suppressed, and signs of robot failure can be detected more effectively. [Brief explanation of the drawing]

[0012] [Figure 1] A perspective view showing the configuration of a robot according to the first embodiment of this disclosure. [Figure 2] A block diagram illustrating the schematic electrical configuration of the robot and condition monitoring device. [Figure 3] A graph showing an example of current time series data. [Figure 4] A graph showing the relationship between encoder temperature and grease temperature. [Figure 5] A conceptual diagram illustrating the corrections applied to the comparison data. [Figure 6] A graph showing the changes in I2 and temperature over time. [Figure 7] A graph showing the changes in I2 and temperature over time. [Figure 8] A graph showing the change in the corrected I2 over time. [Figure 9] This figure shows an example of a screen displayed on the display unit to predict the failure date based on the corrected I2 trend. [Figure 10] A graph illustrating how the prediction line for determining the failure date changes in conjunction with the number of reference days. [Figure 11] A flowchart illustrating the processes performed in a condition monitoring device. [Figure 12] A block diagram schematically showing the electrical configuration of the condition monitoring device in the second embodiment. [Figure 13] A flowchart illustrating the processes performed in the condition monitoring device of the second embodiment. [Figure 14] A conceptual diagram illustrating the processing performed on the comparison data before correction in the fifth embodiment. [Figure 15] A conceptual diagram illustrating the processing performed on the comparison data before correction in the sixth embodiment. [Figure 16] A conceptual diagram illustrating the DTW method used in the seventh embodiment. [Figure 17] A graph showing the temperature dependence of current time series data, relating to the eighth embodiment. [Figure 18] A graph showing the change in motor speed, corresponding to Figure 17. [Modes for carrying out the invention]

[0013] Next, embodiments of the disclosed model will be described with reference to the drawings. Figure 1 is a perspective view showing the configuration of robot 1 according to the first embodiment of this disclosure. Figure 2 is a block diagram showing the electrical configuration of robot 1 and condition monitoring device 5.

[0014] The condition monitoring device 5 relating to this disclosure is applied, for example, to a robot (industrial robot) 1 as shown in Figure 1. The robot 1 performs tasks such as painting, cleaning, welding, and transporting on a workpiece. The robot 1 is implemented, for example, by a vertical articulated robot.

[0015] The configuration of robot 1 will be briefly explained below with reference to Figures 1 and 2, etc.

[0016] Robot 1 comprises a swivel base 10, a multi-joint arm 11, and a wrist unit 12. The swivel base 10 is fixed to the ground (for example, the floor of a factory). The multi-joint arm 11 has multiple joints. The wrist unit 12 is attached to the end of the multi-joint arm 11. An end effector 13 for performing work on a workpiece is attached to the wrist unit 12.

[0017] As shown in Figure 2, the robot 1 is equipped with an arm drive device 21.

[0018] These drive systems consist of actuators, such as servo motors, and reduction gears. However, the configuration of the drive systems is not limited to the above. Each actuator is electrically connected to the controller 90. The actuators operate in accordance with the command values ​​input from the controller 90.

[0019] The driving force from each servo motor constituting the arm drive unit 21 is transmitted via a reduction gear to each joint of the articulated arm 11, the swivel base 10, and the wrist section 12. Each servo motor is fitted with an encoder (not shown) to detect its rotational position.

[0020] Robot 1 performs tasks by reproducing actions recorded through teaching. Controller 90 controls the actuators so that Robot 1 reproduces a series of actions previously taught by the teacher. Teaching Robot 1 can be done by the teacher operating a teaching pendant (not shown in the figure).

[0021] A program for controlling robot 1 is generated through teaching. A new program is generated each time robot 1 is taught. Different actions taught to robot 1 result in different programs. By switching between multiple programs, the actions robot 1 performs can be changed.

[0022] The controller 90 is configured as a known computer, for example, equipped with a CPU, ROM, RAM, auxiliary storage device, etc. The auxiliary storage device is configured as, for example, an HDD, SSD, etc. The auxiliary storage device stores programs for operating the robot 1, etc.

[0023] As shown in Figure 1, the status monitoring device 5 is connected to the controller 90. The status monitoring device 5 acquires the changes in the current value of the current flowing through the actuator (servo motor) via the controller 90.

[0024] If a malfunction occurs in the servo motor or the gearbox connected to it, the servo motor's current value is expected to fluctuate as a result. Therefore, this current value corresponds to a status signal that reflects the state of robot 1. The change in the current value can be represented by repeatedly acquiring the current value at short time intervals and arranging a large number of current values ​​in a time series. Hereinafter, the data obtained by arranging the status signal values ​​in a time series may be referred to as time series data.

[0025] The condition monitoring device 5 can determine whether or not there is an abnormality in the robot 1 by monitoring the acquired time-series data. In this embodiment, the condition monitoring device 5 mainly targets the servo motors and reducers of each joint to determine whether or not there is an abnormality. Here, "abnormality" includes cases where there is some kind of situation in the servo motor, reducer, or bearing that does not lead to malfunction or inability to operate, but is a precursor to such a situation.

[0026] As shown in Figure 2, the condition monitoring device 5 includes a time-series data acquisition unit 51, a basic data acquisition unit (current-related basic data acquisition unit) 52, a basic temperature acquisition unit 53, a comparison data acquisition unit (current-related data acquisition unit) 55, a temperature acquisition unit 56, a correction unit 57, a robot condition evaluation unit 59, and a display unit 60.

[0027] The status monitoring device 5 is configured as a known computer equipped with a CPU, ROM, RAM, auxiliary storage device, etc. The auxiliary storage device is configured as, for example, an HDD or SSD. The auxiliary storage device stores programs for evaluating the status of the robot 1, etc. Through the cooperation of this hardware and software, the computer can be operated as a time-series data acquisition unit 51, a basic data acquisition unit 52, a basic temperature acquisition unit 53, a comparison data acquisition unit 55, a temperature acquisition unit 56, a correction unit 57, a robot status evaluation unit 59, and a display unit 60, etc.

[0028] The time-series data acquisition unit 51 acquires the time-series data described above. The time-series data acquisition unit 51 acquires time-series data for all the servo motors provided in the arm drive device 21 of the robot 1. The time-series data is acquired individually for each of the multiple servo motors (in other words, multiple speed reducers) located in various parts of the robot 1.

[0029] In this embodiment, the status signal is the current value. Here, the current value refers to the measured value of the magnitude of the current flowing through the servo motor, measured by a sensor. The sensor is provided in a servo driver (not shown) that controls the servo motor. However, the sensor may be provided separately from the servo driver for monitoring purposes. Alternatively, the current command value that the servo driver gives to the servo motor may be used as the status signal. The servo driver uses feedback control to bring the servo motor closer to the current command value. Therefore, for the purpose of detecting abnormalities in the servo motor or reduction gear, there is almost no difference between the current value and the current command value.

[0030] The torque of a servo motor is proportional to the current. Therefore, the torque value or torque command value may be used as the status signal.

[0031] As a status signal, the deviation (rotational position deviation) between the target value for the rotational position of the servo motor and the actual rotational position obtained by the encoder may be used. Typically, the servo driver multiplies this deviation by a gain and provides the servo motor with a current command value. Therefore, the change in rotational position deviation shows a similar trend to the change in the current command value.

[0032] The time-series data acquisition unit 51 acquires time-series data for each servo motor each time the robot 1 reproduces a taught action. However, instead of acquiring time-series data for all reproduced actions, it may be possible to acquire data only for, for example, one or a few reproduced actions per day.

[0033] The time-series data acquisition unit 51 acquires time-series data of the current flowing through each servo motor between the timing of receiving the acquisition start signal and the timing of receiving the acquisition end signal. These acquisition start and acquisition end signals are output, for example, by the controller 90.

[0034] The graph in Figure 3 shows an example of the current value flowing to a servo motor of a certain joint when robot 1 performs a regeneration operation. As shown in Figure 3, the current value of the servo motor is zero before the regeneration operation program is executed. At this time, the posture of the multi-joint arm 11, etc., is maintained because electromagnetic brakes (not shown) are operating at each joint.

[0035] Next, the program for robot 1's regeneration operation is started. The brakes are released, and almost simultaneously, current begins flowing to the servo motor. At this point, the servo motor's output shaft is controlled to remain stationary. After a certain amount of time has elapsed, necessary for the angle of the servo motor's output shaft to stabilize, the servo motor begins to rotate. This effectively initiates the operation of robot 1.

[0036] The controller 90 outputs an acquisition start signal to the status monitoring device 5 (and consequently the time-series data acquisition unit 51) shortly after the brake is released and before the servo motor starts rotating.

[0037] Once the sequence of actions taught to robot 1 is complete, the servo motor is controlled to stop rotating. After the servo motor has stopped rotating, and before the program ends, the controller 90 outputs an acquisition completion signal to the time-series data acquisition unit 51.

[0038] In this embodiment, the time-series data is a collection of numerous current values ​​obtained by repeatedly detecting them at short, fixed time intervals, arranged in chronological order. Therefore, the current values ​​in the time-series data are sampled values. The time interval for detecting the current values ​​(sampling interval) is, for example, a few milliseconds. The sampling interval for the current values ​​may or may not coincide with the control cycle of robot 1. In the graph of Figure 3, the time-series data corresponds to the change in current values ​​from the timing of the acquisition start signal to the timing of the acquisition end signal.

[0039] Furthermore, the acquisition start signal or acquisition end signal does not need to be specially prepared for measurement such as the acquisition of time-series data, but can be substantially implemented using other existing signals. For example, the falling edge of the "regeneration in progress" signal, which is mainly used for safety purposes, can be used as the acquisition end signal.

[0040] The time-series data acquisition unit 51 repeatedly acquires time-series data over a long period of time as the robot 1 operates in a factory or the like.

[0041] The basic data acquisition unit 52 obtains two basic data based on the time-series data acquired by the time-series data acquisition unit 51. The two basic data correspond to two different temperatures. The two basic data and the two corresponding different temperatures form the basis of temperature compensation information. The two basic data and the two corresponding different temperatures are acquired for a certain period of time when the robot 1 is considered new and normal. This period can also be called the learning period. The time when the robot 1 is considered new and normal typically means immediately after installation, but it may also be after some trial operation or break-in operation.

[0042] The base temperature acquisition unit 53 acquires the temperature corresponding to the two base data points.

[0043] The following is a detailed explanation. Specifically, after teaching the robot 1 an operation, the time-series data acquisition unit 51 obtains multiple time-series data by having the robot 1 perform the operation multiple times, including the first time. The number of times the operation is performed is arbitrary. When acquiring time-series data, the temperature at the time of acquisition is also acquired. The temperature can be detected, for example, by a temperature sensor (not shown) provided on the encoder described above.

[0044] As will be described later, the no-load running torque, which is the torque loss of the gearbox, changes with temperature, but this temperature can be considered to be the temperature of the grease, which is the lubricant inside the gearbox. The gearbox is usually located on the output shaft side of the motor, and the motor case and the gearbox case are often in contact. The temperature of the grease is affected by the ambient temperature as well as the motor temperature. On the other hand, the encoder is located on the non-output shaft side of the motor (opposite the output shaft), and the encoder case is often in contact with the motor case. The temperature of the encoder is also affected by the ambient temperature as well as the motor temperature. Figure 4 is a graph showing the results of the inventor's verification of the relationship between encoder temperature and grease temperature. Here, the vertical axis shows the grease temperature, and the horizontal axis shows the encoder temperature. Here, a durability test was conducted by having the robot repeatedly perform a certain operation. The graph shows two test results (diamond plot and square plot) when the robot was repeatedly performed different operations. From the linear correlation between the two, the grease temperature can be estimated from the detected encoder temperature. It is also possible to evaluate by considering the encoder temperature as the grease temperature.

[0045] The temperature sensor may be located in a different place from the encoder. For example, a room temperature sensor that measures the room temperature of the factory where robot 1 operates can be used as the temperature sensor. The measurement value from one room temperature sensor can be shared by multiple servo motors.

[0046] In this embodiment, the temperature detected by the temperature sensor is handled in units of 1°C, for example, by truncating the decimal part. However, it is not limited to this, and may be handled in units of 0.5°C, for example.

[0047] The temperature at which time-series data is acquired fluctuates due to various factors such as time, weather, and season. For example, consider a case where 60 time-series data points are obtained through 60 playback operations, including the initial one, with 5 of these points obtained at a temperature of 25°C and another 5 at a temperature of 45°C. The 60 playback operations, including the initial one, refer to the learning period for temperature compensation mentioned above.

[0048] In this example, the basic data acquisition unit 52 of this embodiment obtains first basic data by averaging five time-series data corresponding to a temperature of 25°C. The basic data acquisition unit 52 also obtains second basic data by averaging five time-series data corresponding to a temperature of 45°C. Both the first basic data and the second basic data are time-series data of average values.

[0049] Hereinafter, the temperature at the time of acquiring the first base data will be referred to as the first base temperature, and the temperature at the time of acquiring the second base data will be referred to as the second base temperature. The base temperature acquisition unit 53 acquires the first base temperature (25°C) and the second base temperature (45°C).

[0050] The first basic data corresponds to the first current-related basic data, and the second basic data corresponds to the second current-related basic data.

[0051] The basic data acquisition unit 52 can use a single time series data as the first basic data, instead of using time series data of the average value. However, time series data of current values ​​contain a lot of noise. Therefore, from the viewpoint of improving reliability, it is preferable to use the average of multiple time series data as the first basic data. Similarly, it is preferable to use the average of multiple time series data for the second basic data as well. In the example above, the average of five time series data is calculated, but it is not limited to five; the number of time series data can be any number. From the viewpoint of reliability, it is preferable to have a larger number of time series data to be averaged.

[0052] The first and second base temperatures are determined appropriately so that the temperature difference between them is not too small, and so that a sufficient number of time-series data are obtained for both the first and second base temperatures.

[0053] The first and second basic data may be subjected to known filtering processes.

[0054] The comparison data acquisition unit 55 acquires comparison data. The time series data that forms the basis of the first basic data and the second basic data is acquired only initially, but the time series data acquired initially and thereafter is used as comparison data. In the above example, the repeated regeneration operation performed by the robot 1 can be divided into 60 regeneration operations, including the first one, and subsequent regeneration operations. The basic data acquisition unit 52 acquires the first basic data and the second basic data from the time series data of the initial 60 operations. The comparison data acquisition unit 55 acquires comparison data from the initial 60 operations and the subsequent regeneration operations.

[0055] For comparison data, you can use the time series data as is, or you can use filtered time series data as comparison data.

[0056] Basic data and comparison data are prepared for each playback operation of Robot 1 (in other words, for each program). Basic data and comparison data may be acquired for all taught operations, but for example, if there is an operation that Robot 1 is to perform as the main operation, or a simple movement that is not the main operation but is played back once a day, basic data and comparison data may be acquired only for this operation.

[0057] The temperature acquisition unit 56 acquires the temperature at the time of data acquisition when the comparison data acquisition unit 55 acquires comparison data. The temperature can be acquired in the same way as the base temperature described above. Hereinafter, the temperature at which the comparison data is acquired may be referred to as the data acquisition temperature.

[0058] Filtering of the base data and comparison data is performed, for example, to remove noise contained in time-series data. Since data filtering is publicly known, a detailed explanation will be omitted, but any filter can be used, such as a moving average filter or a CR circuit simulation filter.

[0059] The correction unit 57 corrects the comparison data for the effects of temperature changes.

[0060] The correction is performed as follows. In the condition monitoring device 5 of this embodiment, a predetermined evaluation standard temperature is defined to serve as the basis for data comparison. If the temperature at which the comparison data is acquired differs from the evaluation standard temperature, the correction unit 57 corrects the comparison data so that it corresponds to the data at the evaluation standard temperature. The first basic data, first basic temperature, second basic data, and second basic temperature described above are used for this correction.

[0061] For example, consider a case where the first base temperature is 25°C, the second base temperature is 45°C, the data acquisition temperature is 32°C, and the evaluation standard temperature is 30°C.

[0062] The correction unit 57 calculates the value of the temperature ratio R2, which is defined as R2 = (evaluation reference temperature - data acquisition temperature) / (second base temperature - first base temperature). Substituting the above temperature values, R2 = -0.1.

[0063] The temperature ratio R² is a ratio that indicates how close the data acquisition temperature is to the evaluation reference temperature. If the absolute value of the difference between the data acquisition temperature and the evaluation reference temperature is small, the temperature ratio R² will be close to 0. If the absolute value of the difference is large, the temperature ratio R² will be a value that deviates from 0 in the positive or negative direction. In addition, the temperature ratio R² indicates the relative magnitudes of the data acquisition temperature and the evaluation reference temperature. If the evaluation reference temperature is higher than the data acquisition temperature, R² will be positive, and if the evaluation reference temperature is lower than the data acquisition temperature, R² will be negative.

[0064] The temperature ratio R² is used to correct the comparative data. Therefore, the temperature ratio R² can also be called the temperature ratio for correction.

[0065] As conceptually shown in Figure 5, the correction unit 57 multiplies the difference obtained by subtracting the first basic data from the second basic data by the temperature ratio R2, and adds the resulting result to the comparison data. This obtains the corrected comparison data.

[0066] The corrected comparison data represents the time series data estimated assuming that the data was acquired at the evaluation reference temperature rather than the data acquisition temperature. This correction is performed linearly, as described above, by considering the difference between the temperature at the time the comparison data was acquired (data acquisition temperature) and the evaluation reference temperature.

[0067] The robot state evaluation unit 59 evaluates the state of each part of the robot 1 based on the comparison data corrected by the correction unit 57.

[0068] In this embodiment, the robot state evaluation unit 59 calculates statistics based on the corrected comparison data (time-series data). The statistics can include I2, the maximum current value, the minimum current value, PTP, and PEAK. I2 is the root mean square value of the current. PTP is an abbreviation for Peak to Peak, and is the value obtained by subtracting the current value of the lowest peak from the current value of the highest peak in the current waveform. PEAK is the larger of the absolute values ​​of the maximum current value and the absolute value of the minimum current value. For example, if the maximum current value in the current waveform is +15A and the minimum current value is -12A, then PEAK is 15A. If the maximum current value in the current waveform is +11A and the minimum current value is -14A, then PEAK is 14A.

[0069] Next, we will explain the effect of temperature changes on the statistics of time series data.

[0070] Figure 6 shows the changes in the I2 value and temperature for one of the servo motors of robot 1 in the first example. Figure 7 shows the changes in the I2 value and temperature for the second example. In both graphs, the horizontal axis represents time (specifically, the number of days elapsed). The two data sets were acquired simultaneously, and the temperature changes are identical in both graphs.

[0071] In Figures 6 and 7, it can be seen that the temperature fluctuates gently and periodically. Regarding the calculated I2 value for the first example waveform, as shown in Figure 6, no particular temperature dependence could be observed. Regarding the calculated I2 value for the second example waveform, as shown in Figure 7, a temperature dependence is evident. Figure 8 shows the calculated I2 value for the second example waveform after applying the above correction. In the graph of Figure 8, the I2 value remains stable, indicating that the correction is effective.

[0072] The robot state evaluation unit 59 outputs the obtained statistics such as I2. The robot state evaluation unit 59 can plot the statistics such as I2 over time and calculate a trend line showing the trend. A trend line is a well-known concept and will be briefly explained, but it can be obtained as a straight line that approximates the point cloud plotted on the graph. The approximate straight line can be obtained using the well-known least squares method, but the method is not limited.

[0073] The robot state evaluation unit 59 calculates the date and time when the trend line obtained as described above reaches a predetermined life threshold. By calculating the period from the present to that date and time, the robot state evaluation unit 59 can predict the remaining life of the servo motor, etc.

[0074] The display unit 60 is configured as, for example, a liquid crystal display. The display unit 60 can display various information, such as the changes in statistical quantities like I2 over time, and trend lines.

[0075] The following describes a method for predicting robot failures using a temperature-corrected graph, similar to that in Figure 8, with reference to Figure 9. While Figure 8 showed no increasing trend in the current value I2 monitor due to degradation, the example graph in Figure 9 shows an increasing trend in the current value I2 monitor due to degradation. In the graph in Figure 9, the vertical axis scale is set to be large to make the changes easier to understand.

[0076] Figure 9 shows an example of output from the condition monitoring device 5 to the display unit 60. Based on the past trend of the evaluation value obtained from the current value time series data, the condition monitoring device 5 predicts the future pattern of changes in the evaluation value and displays it on the display unit 60, and also determines the failure date of the robot 1 based on the future pattern of changes in the evaluation value.

[0077] To support maintenance, the condition monitoring device 5 can display a trend management screen on the display unit 60, as shown in Figure 9. Therefore, the condition monitoring device 5 also functions as a maintenance support device. In the example in Figure 9, the trend management screen includes a graph display unit 61. The graph display unit 61 shows the trend of the current value I2 monitor in a graph. This current value I2 is corrected for the effects of temperature changes.

[0078] The status monitoring device 5 can change the display mode of the graph displayed on the graph display unit 61 as shown in Figure 10. In Figure 10, one end of the horizontal axis of the graph display unit 61 is the start date of drawing, and the other end of the horizontal axis is the end date of drawing. The period from the start date of drawing to the end date of drawing corresponds to the display period of evaluation values ​​and prediction lines, etc. In the example in Figure 10, the start date of drawing and the end date of drawing can be changed. For example, if the start date of drawing is set to 1 number of days before the current day and the end date of drawing is set to 2 number of days after the current day, the operator can change these 1st and 2nd number of days independently. This allows the operator to grasp evaluation values ​​within the range they desire.

[0079] As mentioned above, the corrected comparison data is time-series data estimated assuming it was acquired at the evaluation standard temperature. Therefore, the effects of temperature changes can be eliminated. As a result, the robot state evaluation unit 59 can evaluate the state of the robot 1 more appropriately.

[0080] In the example above, the evaluation reference temperature is set to be different from both the first and second base temperatures. However, the evaluation reference temperature may be equal to either the first or second base temperature.

[0081] Next, the processing of the status monitoring device 5 will be explained with reference to Figure 11.

[0082] When the process shown in Figure 11 starts, the time-series data acquisition unit 51 of the condition monitoring device 5 acquires current time-series data each time the robot 1 performs a regeneration operation, and saves it along with the acquisition date and time and temperature information (step S101). The condition monitoring device 5 repeats the above process until a predetermined period of time has elapsed (step S102).

[0083] When a predetermined period has elapsed, the base temperature acquisition unit 53 of the condition monitoring device 5 acquires the lowest temperature for which, for example, five or more current time-series data have been stored, as the first base temperature. The base data acquisition unit 52 then calculates the average value time-series data based on the multiple current time-series data acquired at this first base temperature (step S103). This average value time-series data is the first base data.

[0084] Next, the base temperature acquisition unit 53 acquires the highest temperature for which, for example, five or more current time series data are stored, as the second base temperature. The base data acquisition unit 52 then calculates the average value time series data based on the multiple current time series data acquired at this second base temperature (step S104). This average value time series data is the second base data.

[0085] Steps S101 to S104 constitute the preparation process, i.e., the learning period for temperature compensation. Next, the process of actually evaluating the state of robot 1 begins.

[0086] First, the comparison data acquisition unit 55 acquires time-series data from the time-series data acquisition unit 51. This time-series data is the comparison data. At this time, the temperature acquisition unit 56 acquires the temperature at the time the comparison data was acquired. This temperature is the data acquisition temperature.

[0087] Once comparison data and data acquisition temperatures are obtained, the correction unit 57 calculates the temperature ratio R2 based on the difference between the evaluation reference temperature and the data acquisition temperature. The correction unit 57 multiplies the difference obtained by subtracting the first basic data from the second basic data by the temperature ratio R2. As a result, time-series data for correcting the comparison data can be obtained (step S105).

[0088] Next, the correction unit 57 corrects the comparison data by adding the above time-series data to the comparison data (step S106).

[0089] The robot state evaluation unit 59 calculates statistical quantities such as I2 for the comparison data corrected in step S106 and outputs them to the display unit 60 (step S107).

[0090] After that, the process returns to step S105, and the processes from step S105 to S107 are repeated.

[0091] As described above, the state monitoring device 5 of this embodiment monitors the state of the robot 1 capable of reproducing predetermined operations. The state monitoring device 5 comprises a basic data acquisition unit 52, a basic temperature acquisition unit 53, a comparison data acquisition unit 55, a temperature acquisition unit 56, a correction unit 57, and a robot state evaluation unit 59. The basic data acquisition unit 52 acquires first basic data, which is current data, and second basic data, which is current data, for the current of the motor that drives the robot 1. The basic temperature acquisition unit 53 acquires first basic temperature, which is the temperature at the time of acquisition of first basic data, and second basic temperature, which is the temperature at the time of acquisition of second basic data. The comparison data acquisition unit 55 acquires comparison data, which is current data, for the motor current. The temperature acquisition unit 56 acquires data acquisition temperature, which is the temperature at the time of acquisition of comparison data. The correction unit 57 performs corrections on the comparison data based on the first basic data, second basic data, first basic temperature, second basic temperature, and data acquisition temperature. The robot state evaluation unit 59 evaluates the state of robot 1 using the corrected comparison data.

[0092] This allows for the evaluation of the robot's state while suppressing the effects of temperature changes through correction. Consequently, early signs of robot failure can be detected effectively.

[0093] In the condition monitoring device 5 of this embodiment, the robot condition evaluation unit 59 calculates I2, the maximum value of the current, the minimum value of the current, PTP, or PEAK based on the corrected comparison data, and performs a robot condition evaluation based on these calculation results.

[0094] This allows for evaluation using statistical measures while eliminating the effects of temperature changes.

[0095] In the condition monitoring device 5 of this embodiment, the first basic data, the second basic data, and the comparison data are time-series data of current.

[0096] This allows for a more accurate evaluation of the state of robot 1 using detailed data.

[0097] In the condition monitoring device 5 of this embodiment, the correction unit 57 corrects the comparison data so that it corresponds to the data at a predetermined evaluation reference temperature.

[0098] This enables evaluation at a consistent temperature.

[0099] In the state monitoring device 5 of this embodiment, the correction unit 57 performs a linear correction on the current-related data.

[0100] This allows us to suppress the effects of temperature changes with simple calculations.

[0101] In the status monitoring device 5 of this embodiment, the first basic data and the second basic data are time-series data of average values ​​based on a plurality of time-series data.

[0102] This can improve the reliability of status monitoring.

[0103] Next, a second embodiment will be described. In the description of this embodiment and subsequent embodiments, the same or similar components as those in the previously described embodiments will be denoted by the same reference numerals in the drawings, and their descriptions may be omitted.

[0104] In this embodiment, as shown in Figure 12, the status monitoring device 5 includes a reference data acquisition unit 54 in addition to the configuration of the first embodiment.

[0105] The reference data acquisition unit 54 determines the reference data based on the two basic data obtained by the basic data acquisition unit 52.

[0106] Reference data refers to time-series data corresponding to the evaluation standard temperature mentioned above. More specifically, the reference data is estimated by interpolation or extrapolation of time-series data at the evaluation standard temperature, based on the first and second basic data. Here, we consider the case where the evaluation standard temperature is 30°C.

[0107] The reference data acquisition unit 54 calculates the value of the temperature ratio R1, which is defined as R1 = (evaluation reference temperature - first base temperature) / (second base temperature - first base temperature). In this example, the first base temperature is 25°C, the second base temperature is 45°C, and the evaluation reference temperature is 30°C. Substituting these values, R1 = 0.25.

[0108] The temperature ratio R1 is a ratio that indicates how close the evaluation reference temperature is to the first base temperature. If the evaluation reference temperature is close to the first base temperature, the temperature ratio R1 will be close to 0, and if the evaluation reference temperature is close to the second base temperature, the temperature ratio R1 will be close to 1.

[0109] The reference data acquisition unit 54 subtracts the first basic data from the second basic data, multiplies the difference obtained by the above-mentioned temperature ratio R1, and adds the resulting result to the first basic data. Thus, the reference data is obtained.

[0110] Here, we consider a multidimensional space with a number of dimensions equal to the number of elements in the time-series data. The first and second basic data are plotted in this multidimensional space. The reference data corresponds to a point obtained by dividing the line segment connecting the points of the first and second basic data in the multidimensional space. The division ratio in this case is determined based on the ratio indicating how close the evaluation reference temperature is to the first and second basic temperatures.

[0111] The above is an example of obtaining reference data by interpolation, but there may be cases where the evaluation reference temperature is not between the first base temperature and the second base temperature. In this case, the reference data can be obtained by externally dividing the line segment connecting the points of the two base data points, rather than internally. Thus, reference data can also be obtained by extrapolation.

[0112] The reference data obtained in the manner described above refers to time-series data estimated under the assumption that it was acquired at the evaluation reference temperature, not at the first or second base temperature.

[0113] Basic data and reference data are prepared for each playback operation of Robot 1 (in other words, for each program). Basic data and reference data may be acquired for all taught operations, but for example, if there is an operation that Robot 1 is to perform as the main operation, or a simple movement that is not the main operation but is played back once a day, basic data and reference data may be acquired only for this operation.

[0114] In this embodiment, the robot state evaluation unit 59 calculates a Euclidean distance (dissimilarity) that indicates the magnitude of the difference between the corrected comparison data and the reference data. The robot state evaluation unit 59 outputs the obtained Euclidean distance as an evaluation quantity.

[0115] The Euclidean distance D between time series data a and time series data b is expressed by the following equation (1), where m is the length (number of elements) of the time series data.

number

[0116] Since the Euclidean distance is publicly known, a detailed explanation will be omitted.

[0117] The robot state evaluation unit 59 evaluates the state of each part of the robot 1 using the Euclidean distance calculated as described above. Specifically, the robot state evaluation unit 59 evaluates the state of each servo motor using the Euclidean distance acquired for each servo motor.

[0118] As time progresses, it is expected that the waveform corresponding to the time-series data of the current value will diverge from the waveform corresponding to the time-series data of the initially obtained current value. The Euclidean distance can be considered a numerical representation of the degree of this divergence. The robot state evaluation unit 59 uses the Euclidean distance to determine whether or not there is an abnormality in each of the six joint servo motors. Furthermore, based on the change in the Euclidean distance over time up to the present, it is possible to predict the timing at which the servo motors will malfunction or become inoperable in the future.

[0119] The robot state evaluation unit 59 can calculate a trend line showing the trend of Euclidean distance over time. The robot state evaluation unit 59 calculates the date and time when the trend line obtained as described above reaches a predetermined life threshold. By calculating the period from the present to that date and time, the robot state evaluation unit 59 can predict the remaining life of the servo motor, etc.

[0120] The display unit 60 is configured as, for example, a liquid crystal display. The display unit 60 can display various information, such as the change in the Euclidean distance over time and trend lines.

[0121] As mentioned above, the reference data is time-series data estimated assuming it was acquired at the evaluation reference temperature. The same applies to the corrected comparison data. Therefore, the effect of temperature changes can be removed when calculating the dissimilarity. As a result, the robot state evaluation unit 59 can evaluate the state of the robot 1 more appropriately.

[0122] In the example above, the evaluation reference temperature is set to be different from both the first and second base temperatures. However, the evaluation reference temperature may be equal to either the first or second base temperature. In this case, the value of the temperature ratio R1 mentioned above becomes 0 or 1, which simplifies the calculation.

[0123] The following describes the processing of the status monitoring device 5 with reference to the flowchart in Figure 13. Steps S201 to S204 in Figure 13 are the same as steps S101 to S104 in Figure 11, so their explanation will be omitted.

[0124] In this embodiment, the processing in step S205 is performed after step S204 and before step S206. In step S205, the reference data acquisition unit 54 interpolates or extrapolates the first basic data and the second basic data based on the aforementioned evaluation reference temperature to obtain time series data. This time series data is the reference data.

[0125] Steps S206 and S207 in Figure 13 are the same as steps S105 and S106 in Figure 11, respectively, so their explanation is omitted.

[0126] In step S208, the robot state evaluation unit 59 calculates the dissimilarity between the reference data obtained in step S205 and the comparison data corrected in step S207. The obtained dissimilarity is output to the display unit 60.

[0127] The process then returns to step S206, and steps S206 through S208 are repeated.

[0128] As described above, in the state monitoring device 5 of this embodiment, reference data to be compared with the comparison data is obtained based on the first basic data and the second basic data. The robot state evaluation unit 59 determines the dissimilarity between the comparison data and the reference data.

[0129] This allows for the appropriate evaluation of the degree of change in the current waveform while suppressing the effects of temperature changes.

[0130] In the state monitoring device 5 of this embodiment, the robot state evaluation unit 59 determines the dissimilarity by calculating the Euclidean distance between the comparison data and the reference data.

[0131] This allows for a precise evaluation of the degree of change in the current waveform.

[0132] In the condition monitoring device 5 of this embodiment, the correction unit 57 performs linear correction by adding to the comparison data the difference obtained by subtracting the first basic data from the second basic data, multiplied by (evaluation reference temperature - data acquisition temperature) / (second basic temperature - first basic temperature).

[0133] This allows for a simplification of calculations.

[0134] Next, a third embodiment will be described.

[0135] In the first embodiment described above, no reference data is created, and statistics such as I2 are calculated using corrected comparison data, and the state of robot 1 is evaluated based on these statistics. In the second embodiment, reference data is created, and the similarity between the reference data and the corrected comparison data is considered, and the state of robot 1 is evaluated using the Euclidean distance (dissimilarity).

[0136] In this embodiment and subsequent embodiments, with some exceptions, any of the above evaluation methods can be applied. When the description states "when creating reference data," it means that evaluation can be done not only using statistical measures, but also using dissimilarity measures such as Euclidean distance between the reference data and the comparison data.

[0137] In this embodiment, when calculating the reference data, interpolation or extrapolation is not simply performed on the first basic data and the second basic data using the temperature ratio, but rather interpolation or extrapolation is performed using the current ratio obtained from the temperature ratio.

[0138] The following explains in detail. Assuming that a large portion of the motor's lost torque is accounted for by the no-load running torque, the current corresponding to this lost torque is, for example, tp U It has the property of becoming approximately 1 / A for every increase in °C, where A is a number greater than 1. Details will be explained later.

[0139] Therefore, for every change in Δtp℃, the current is A^(-Δtp / tp U Reference data can be obtained by interpolating or extrapolating the first and second base data using a function that represents change. The above function ΔI=f(Δtp) is also used when correcting comparison data.

[0140] It is also possible to use the above function ΔI=f(Δtp) solely to correct the comparison data without creating reference data. In this case, I2, etc., are calculated from the corrected comparison data and used for evaluation.

[0141] As explained above, in the state monitoring device 5 of this embodiment, the function ΔI = f(Δtp) = A^(-Δtp / tp) expresses the current change ΔI in terms of the temperature change Δtp. U ) is defined. The correction unit 57 performs a linear correction by adding to the current-related data the difference obtained by subtracting the first basic data from the second basic data multiplied by f(evaluation reference temperature - data acquisition temperature) / f(second basic temperature - first basic temperature).

[0142] This allows for correction of the comparison data in a way that more appropriately reflects the nature of current changes in response to temperature changes.

[0143] Next, a fourth embodiment will be described. In the fourth embodiment, the correction unit 57 performs correction on the comparison data in a different manner than in the above embodiments.

[0144] In the example above, the first and second basic data, which are current time series data, are acquired at different temperatures. If there is a difference in current values ​​between the first and second basic data, this difference in current values ​​corresponds to a combination of the fluctuations in dry friction torque and the fluctuations in no-load running torque.

[0145] The dry friction torque refers to the lost torque when the rotational speed of the motor is near zero, and means the lost torque due to known dry friction. However, the torque when the rotational speed of the motor is near zero is considered to include lost torques other than the dry friction torque, such as hysteresis loss. The no-load running torque means the lost torque due to the viscosity of the lubricant when the motor is rotated under no load. Hereinafter, the current value corresponding to the sum of the dry friction torque and the no-load running torque may be referred to as the lost torque current value.

[0146] The lost torque current value has the property of becoming 1 / A times each time the temperature increases by a predetermined unit temperature t U . However, A is a number greater than 1. The combination of the values of A and t U can be arbitrarily determined. For example, when t U is 10 °C, depending on the type of the speed reducer, etc., the value of A can be determined to be 1.1 or more and 1.6 or less. For example, the value of A may be changed between the high temperature range and the low temperature range.

[0147] Considering the above properties, when the lost torque current value I lt,tp1 at the first base temperature, the current value of the first base data is I1, the current value of the second base data is I2, the first base temperature is tp1, and the second base temperature is tp2, it can be expressed by the following formula. I lt,tp1 =(I1 - I2) / (1 - A^((tp1 - tp2) / t U ))

[0148] The lost torque current value I X at the data acquisition temperature tp lt,tpX can be expressed by the following formula. I lt,tpX =I lt,tp1 ×A^((tp1 - tp X ) / t U ) The lost torque current value I S at the evaluation reference temperature tp lt,tpS can be expressed by the following formula. I lt,tpS =Ilt,tp1 ×A^((tp1-tp S ) / t U )

[0149] Therefore, the obtained comparative data I ORIG In contrast, data acquisition temperature tp X Lost torque current value I lt,tpX Subtract the following, and further, the evaluation standard temperature tp S Lost torque current value I lt,tnS By adding this, the evaluation standard temperature tp S Comparative data I corrected to correspond to COMP You can obtain this. I COMP =I ORIG -I lt,tpX +I lt,tnS With the above steps, the correction by the correction unit 57 is completed.

[0150] Comparison Data I ORIG Data acquisition temperature tp X Lost torque current value I lt,tpX By simply subtracting, the evaluation standard temperature tp S Lost torque current value I lt,tnS It is also possible to avoid adding (I COMP =I ORIG -I lt,tpX ). In this case, comparison data I ORIG In this case, the current equivalent to the motor's lost torque can be simply canceled. Correspondingly, in the above reference data, the evaluation reference temperature tp S Lost torque current value I lt,tnS This is subtracted. As a result, the reference data and the corrected comparison data can be appropriately compared after removing the current corresponding to the motor's lost torque.

[0151] Data acquisition temperature tp X Lost torque current value I lt,tpX When calculating tp in the above formula, X Everything else is a constant. Therefore, tp X From I lt,tpXCreating a conversion table in advance allows for faster calculations and easier adjustment of fine parameters. (Evaluation standard temperature tp) S Lost torque current value I lt,tpS The same applies when seeking [the result].

[0152] As described above, in the condition monitoring device 5 of this embodiment, the first basic data, the second basic data, and the current-related data are time-series data of current. The correction unit 57 performs corrections on the comparison data with respect to the temperature characteristics of the motor's lost torque.

[0153] This allows for correction that appropriately considers the temperature characteristics of the motor's lost torque.

[0154] In the state monitoring device 5 of this embodiment, the correction unit 57 determines that when A is a number greater than 1, the temperature is a predetermined unit temperature t U The comparison data is corrected using the relationship that the current value corresponding to the lost torque becomes 1 / A for each increase.

[0155] This allows for correction that appropriately considers the temperature characteristics of the lost torque.

[0156] In the state monitoring device 5 of this embodiment, the unit temperature t U When the temperature is set to 10°C, A is between 1.1 and 1.6.

[0157] This allows for corrections that specifically consider the temperature characteristics of the lost torque.

[0158] In the condition monitoring device 5 of this embodiment, the correction unit 57 can also perform corrections on the comparison data so as to cancel the current corresponding to the motor's lost torque.

[0159] In this case, the current corresponding to the motor's lost torque can be removed from the time-series data of the current. Therefore, the evaluation becomes simpler, which may be advantageous for evaluating the state of robot 1.

[0160] Next, a fifth embodiment will be described.

[0161] This embodiment differs from the second embodiment described above in that the robot state evaluation unit 59 performs a process of moving the comparison data relative to the reference data (comparison target data) in the time axis direction in order to match its phase.

[0162] Specifically, the robot state evaluation unit 59 in this embodiment performs a process of moving the waveform of the comparison data in multiple stages along the time axis. For each stage, the robot state evaluation unit 59 calculates and obtains the Euclidean distance (dissimilarity) between the comparison data and the reference data.

[0163] Figure 14 shows an example where the comparison data is shifted by a total of two steps: one step in the negative direction and one step in the positive direction on the time axis. The Euclidean distance is calculated for three steps, including the case with no shift. The amount of shift in the time axis direction per step, and the number of shift steps, can both be arbitrarily determined.

[0164] This waveform shift can be effectively achieved by uniformly shifting the correspondence between m current values ​​in the reference data and m current values ​​in the comparison data when calculating the Euclidean distance.

[0165] In this embodiment, the robot state evaluation unit 59 determines the minimum value of the Euclidean distance. This minimum value of the Euclidean distance is used as the evaluation value.

[0166] In the embodiment described above, if interpolation or extrapolation is simply performed, even a slight time delay between the waveform of the comparison data and the waveform of the reference data will significantly affect the Euclidean distance, i.e., the evaluation value. In this embodiment, however, the comparison data is moved relative to the reference data in the time axis direction to improve the degree of agreement between the comparison data and the reference data. Therefore, an appropriate evaluation can be performed.

[0167] It is preferable to set an upper limit for the amount by which the comparison data is shifted relative to the reference data in the time axis direction. This allows for the ignoring of phase shifts as measurement errors, while appropriately detecting large phase shifts that indicate deterioration of servo motors, etc.

[0168] As described above, in the state monitoring device 5 of this embodiment, the robot state evaluation unit 59 determines the dissimilarity between two waveforms while moving at least one of the waveforms of the reference data and the waveform of the comparison data in multiple steps in the time axis direction. When the comparison data is moved to the state in which the dissimilarity between the two waveforms is minimized, the robot state evaluation unit 59 determines the dissimilarity between the reference data and the comparison data.

[0169] This allows for the evaluation of the dissimilarity between the reference data and the comparison data while correcting for a certain degree of time-axis discrepancy between the reference data and the comparison data. Consequently, the evaluation of the state of robot 1 becomes more appropriate.

[0170] Next, a sixth embodiment will be described.

[0171] In this embodiment, the correction unit 57 calculates the difference between the i-th current value out of m current values ​​included in the reference data and the current values ​​from the ip-th to the i+p-th current values ​​out of m current values ​​included in the comparison data. Figure 15 is a schematic diagram showing the case where p=2, and the difference in current values ​​is represented by the underlined numbers.

[0172] Next, the correction unit 57 checks which of the current values ​​from the ip-th to the i+p-th of the comparison data is closest to the i-th current value of the reference data. In the example in Figure 15, among the current values ​​from the i-2-th to the i+2-th of the comparison data, the one with the smallest absolute difference from the i-th current value of the reference data is the i-1-th, which has a difference of -0.2. The correction unit 57 corrects the i-th current value of the comparison data so that it becomes the i-1-th current value, which has the smallest absolute difference.

[0173] The correction unit 57 repeats the above process for all m current values ​​of the reference data. This correction allows the comparison data to be moved to some extent in the time axis direction and stretched to some extent in the time axis direction in order to increase its similarity to the reference data. Subsequently, the robot state evaluation unit 59 calculates the Euclidean distance between the reference data and the corrected comparison data as the dissimilarity. This enables appropriate evaluation.

[0174] As described above, in the state monitoring device 5 of this embodiment, reference data to be compared with the comparison data is obtained based on at least one of the first basic data and the second basic data. The robot state evaluation unit 59 corrects the i-th sampling value of the comparison data with the sampling value that is closest to the i-th sampling value of the reference data waveform from the sampling values ​​of the comparison data from the ip-th to the i+p-th (where i and p are integers of 1 or more) sampling values ​​of the comparison data. The robot state evaluation unit 59 then determines the dissimilarity between the corrected comparison data and the reference data.

[0175] This allows for the evaluation of the comparison data while correcting, to a certain extent, the time-axis deviation and distortion between the reference data and the comparison data. Consequently, the evaluation of the state of robot 1 becomes more appropriate.

[0176] Next, a seventh embodiment will be described.

[0177] The state monitoring device 5 of this embodiment includes a reference data acquisition unit 54, similar to the second embodiment described above. This embodiment differs from the second embodiment in that the robot state evaluation unit 59 calculates the DTW distance based on the DTW method as a value indicating the dissimilarity between the comparison data and the reference data (comparison target data), and uses this as the evaluation value. DTW is an abbreviation for Dynamic Time Warping.

[0178] Let's briefly explain the DTW method. The DTW method is used to calculate the degree of similarity between two time series data. A key feature of the DTW method is that it allows for non-linear stretching and compression of the time series data along the time axis when calculating similarity. As a result, the DTW method can obtain results regarding the similarity of time series data that are close to human intuition.

[0179] In this embodiment, the robot state evaluation unit 59 calculates the DTW distance (dissimilarity) which indicates the magnitude of the difference between the comparison data and the reference data, and outputs this value as the evaluation value.

[0180] The principle of the DTW method will be explained using Figure 16. Multiple (s) current values ​​included in the reference data are arranged in chronological order along the first axis, which extends horizontally. Multiple (p) current values ​​included in the comparison data are arranged in chronological order along the second axis, which extends vertically.

[0181] Next, we define s × p cells arranged in a matrix on a plane defined by the vertical and horizontal axes. Each cell (c,d) represents the correspondence between the c-th current value in the reference data and the d-th current value in the comparison data, where 1 ≤ c ≤ d.

[0182] Each cell (c,d) is associated with a numerical value representing the difference between the c-th current value in the reference data and the d-th current value in the comparison data. In this embodiment, each cell is associated with and stores the absolute value of the difference between the c-th current value and the d-th current value.

[0183] The robot state evaluation unit 59 determines the warping path (route) from the starting cell located in the lower left corner of the matrix in Figure 16 to the ending cell located in the upper right corner.

[0184] The starting cell (1,1) corresponds to associating the current value at the earliest timing (i.e., the 1st) in the time series among the s current values ​​in the reference data with the current value at the earliest timing (i.e., the 1st) in the time series among the p current values ​​in the comparison data.

[0185] The endpoint cell (s,p) corresponds to associating the current value at the last timing (i.e., the sth) in the time series among the s current values ​​in the reference data with the current value at the last timing (i.e., the pth) in the time series among the p current values ​​in the comparison data.

[0186] In the s × p matrix constructed as described above, consider a path from the starting cell to the ending cell, following the rules [1] and [2] below: [1] Movement is only possible to adjacent cells vertically, horizontally, or diagonally. [2] Movement is not possible in the direction of reversing time for the reference data, nor is it possible to move in the direction of reversing time for the comparison data.

[0187] A sequence of cells like this is called a path or warping path. A warping path shows how s current values ​​in the reference data correspond to p current values ​​in the comparison data. In other words, a warping path represents how two time-series data are stretched or compressed along the time axis.

[0188] Multiple warping paths are possible from the starting cell to the ending cell. The robot state evaluation unit 59 finds the warping path that minimizes the sum of numerical values ​​representing the differences associated with the cells traversed (in this embodiment, the absolute value of the difference between the c-th current value and the d-th current value) among the possible warping paths. Hereafter, this warping path may be referred to as the optimal warping path. The sum of the values ​​within each cell in this optimal warping path may be referred to as the DTW distance. The robot state evaluation unit 59 finds this DTW distance.

[0189] The average DTW distance can be calculated by dividing the DTW distance by the number of cells passed through. Alternatively, the average DTW distance can be calculated by dividing the DTW distance by the number of elements (s or p) in either the time series data. The average DTW distance can also be used as the evaluation value instead of the DTW distance.

[0190] When s and p are large, a vast number of warping paths are possible. Therefore, if we were to consider all possible warping paths, the computational cost required to find the optimal warping path would increase exponentially. To solve this problem, the robot state evaluation unit 59 of this embodiment uses the DP matching method (dynamic programming) to find the optimal warping path. DP is an abbreviation for Dynamic Programming. The DP matching method is well known, so its explanation will be omitted.

[0191] As described above, in the state monitoring device 5 of this embodiment, reference data to be compared with the comparison data is determined based on at least one of the first basic data and the second basic data. The robot state evaluation unit 59 determines the DTW distance between the comparison data and the reference data as the dissimilarity.

[0192] This allows for the acquisition of dissimilarity by effectively excluding the effects of time delays or distortions in the time axis direction between the comparison data and the reference data.

[0193] Next, the eighth embodiment will be described.

[0194] For example, in the first embodiment, the first basic data, the second basic data, and the comparison data are all time-series data of current values, and statistics are calculated from the comparison data. On the other hand, in this embodiment, the first basic data, the second basic data, and the comparison data are data obtained by statistically processing time-series data of current values ​​in an appropriate manner.

[0195] As mentioned above, the statistical data can be any of the following: I2, maximum current, minimum current, PTP, and PEAK.

[0196] When using statistical data, the first base data, second base data, and comparison data are simple scalar quantities. Therefore, calculations for interpolation or extrapolation become simpler when calculating the baseline data and when correcting the comparison data.

[0197] As described above, in the state monitoring device 5 of this embodiment, the first current-related basic data, the second current-related basic data, and the current-related data are I2, the maximum value of the current, the minimum value of the current, PTP, or PEAK.

[0198] This allows the effects of temperature changes on statistical current data to be suppressed through correction.

[0199] Next, the ninth embodiment will be described. First, Figures 6 to 1 described above. 0 We will now explain the effects of temperature changes by referring to a different graph than the one used previously.

[0200] The graph in Figure 17 shows the current waveforms for one of the servo motors on robot 1, obtained by subtracting the current waveform at 40°C from the current waveform at 35°C, and also by subtracting the current waveform at 50°C from the current waveform at 35°C. The vertical axis of the graph represents the current value, and the horizontal axis represents time (specifically, seconds). The fact that the two difference waveforms are clearly different indicates that the current waveform is temperature-dependent.

[0201] The graph in Figure 18 shows the change in servo motor speed when robot 1 is made to perform a regeneration operation using the current waveform corresponding to Figure 17. The vertical axis of the graph represents speed, and the horizontal axis represents time (specifically, seconds). The time on the horizontal axis of Figure 18 corresponds to the time on the horizontal axis of Figure 17. As shown in Figure 18, the period during which robot 1 performs a regeneration operation using the current waveform that forms the basis of the difference waveform in Figure 17 includes a period in which the motor rotation is substantially zero and a period in which it is substantially not zero.

[0202] In the graph in Figure 17, the first period P1, indicated by the white arrow, is the period during which the motor rotation is virtually zero. The current value during this first period P1 is considered to correspond to the dry friction torque. The second period P2, which does not have an arrow, is the period during which the motor rotation is not virtually zero. The current value during this second period P2 is considered to correspond to the no-load running torque.

[0203] If the comparison data is time-series data, for example, the correction unit 57 can perform correction on the comparison data only for either the first period P1 or the second period P2 shown in Figure 17. Furthermore, for example, a simple linear correction as shown in the first embodiment can be applied to the comparison data for the first period P1, while for the second period P2, the temperature can be set to a predetermined unit temperature t. U The comparative data can be corrected using the property that it becomes 1 / A for each increase. Alternatively, a simple linear correction can be performed on both periods, and the correction can be applied to the basic data separated for each period. Furthermore, if the temperature in both periods is a predetermined unit temperature t U By utilizing the property that the value of A multiplies by 1 / A with each increase, it is also possible to correct the comparison data with different values ​​of A for each period.

[0204] In this way, the time-series data of the current value can be divided into periods when the motor rotation is substantially zero and periods when it is not, and the correction method or parameters performed by the correction unit 57 can be different for each period. Alternatively, correction can be performed for only one of the two. This makes it possible to more effectively eliminate the effects of temperature changes.

[0205] As described above, in the state monitoring device 5 of this embodiment, when the comparison data is divided into a first period P1 in which the motor rotation speed is substantially zero and a second period P2 in which the motor rotation speed is substantially not zero, the correction unit 57 performs a correction on only one of the first period P1 and the second period P2. Alternatively, different corrections are performed on the first period P1 and the second period P2.

[0206] This allows for better suppression of the effects of temperature changes.

[0207] Preferred embodiments of the present disclosure have been described above, but the above configuration can be modified as follows, for example. Modifications may be made individually or in any combination of multiple modifications.

[0208] The status monitoring device 5 does not need to be directly connected to the robot 1; for example, it may acquire time-series data reflecting the status of the robot 1 from the controller 90 of the robot 1 via a communication line such as the Internet. In this case, the controller 90 acquires and stores the current value in real time during the regeneration operation, and transmits the time-series data of the current value, along with the program number, the date and time the current value was acquired, and information identifying the servo motor, to the status monitoring device 5 in batch processing or the like.

[0209] The status monitoring device 5 does not have to be provided separately from the controller 90, but may be built into the controller 90. Furthermore, the status monitoring device 5 may not have a separate computer functioning as a CPU, ROM, RAM, auxiliary storage device, etc., and may be implemented using the computer of the robot 1's controller 90.

[0210] The functions of each element, including the condition monitoring device 5 disclosed above, can be performed using circuits or processing circuits, including general-purpose processors, dedicated processors, integrated circuits, ASICs (Application Specific Integrated Circuits), conventional circuits, and / or combinations thereof, configured or programmed to perform the disclosed functions. A processor is considered a processing circuit or circuit because it includes transistors and other circuits. In this disclosure, a circuit, unit, or means is hardware that performs the enumerated functions, or hardware programmed to perform the enumerated functions. The hardware may be hardware disclosed herein, or other known hardware that is programmed or configured to perform the enumerated functions. If the hardware is a processor, which is considered a type of circuit, then the circuit, means, or unit is a combination of hardware and software, and the software is used to configure the hardware and / or the processor.

[0211] From the above disclosure, at least the following technical concepts can be understood.

[0212] (Item 1) A condition monitoring device for monitoring the status of an industrial robot capable of reproducing predetermined actions, A current-related basic data acquisition unit acquires, with respect to the current of the motor that drives the industrial robot, first current-related basic data which is at least one current data or statistical data of the current data, and second current-related basic data which is at least one current data or statistical data of the current data. A base temperature acquisition unit that acquires a first base temperature, which is the temperature at the time of acquiring the first current-related base data, and a second base temperature, which is the temperature at the time of acquiring the second current-related base data, A current-related data acquisition unit acquires current-related data, which is at least one current data or statistical data of the current data, for the current of the motor. A temperature acquisition unit that acquires the data acquisition temperature, which is the temperature at the time of acquiring the current-related data, A correction unit that performs corrections to the current-related data based on the first current-related basic data, the second current-related basic data, the first basic temperature, the second basic temperature, and the data acquisition temperature, A robot state evaluation unit that evaluates the state of the robot using the corrected current-related data, A condition monitoring device equipped with the following features.

[0213] (Item 2) A condition monitoring device as described in Item 1, The robot state evaluation unit, based on the corrected current-related data, The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or The larger of the absolute values ​​of the maximum and minimum current values. A condition monitoring device that calculates a value and evaluates the robot's condition based on this calculation result.

[0214] (Item 3) A condition monitoring device as described in Item 1, Based on at least one of the first current-related basic data and the second current-related basic data, comparison data to be compared with the current-related data is obtained. The robot state evaluation unit is a state monitoring device that determines the degree of dissimilarity between the current-related data and the comparison target data.

[0215] (Item 4) A condition monitoring device as described in Item 1, The robot state evaluation unit is a state monitoring device that determines the dissimilarity by taking the Euclidean distance between the current-related data and the comparison target data.

[0216] (Item 5) A condition monitoring device as described in Item 3 or 4, The robot state evaluation unit determines the degree of dissimilarity between the current-related data and the comparison data while relatively moving the current-related data in multiple steps in the time axis direction relative to the comparison data. A state monitoring device in which, in a state in which the current-related data has been moved relative to itself so that the dissimilarity is minimized, the robot state evaluation unit determines the dissimilarity between the current-related data and the comparison target data.

[0217] (Item 6) A condition monitoring device as described in Item 3, The robot state evaluation unit is a state monitoring device that determines the DTW distance between the current-related data and the comparison target data as the dissimilarity.

[0218] (Item 7) A condition monitoring device described in any one of items 1 to 6, A state monitoring device in which the first current-related basic data, the second current-related basic data, and the current-related data are time-series data of current.

[0219] (Item 8) A condition monitoring device as described in Item 1, The first current-related basic data, the second current-related basic data, and the current-related data are The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or A state monitoring device that uses the larger of the absolute values ​​of the maximum and minimum current values.

[0220] (Item 9) A condition monitoring device as described in any one of items 1 to 8, The correction unit is a condition monitoring device that corrects the current-related data so that it corresponds to data at a predetermined evaluation reference temperature.

[0221] (Item 10) A condition monitoring device described in any one of items 1 to 7, or item 9, The correction unit performs a linear correction by adding to the current-related data the difference obtained by subtracting the first current-related basic data from the second current-related basic data, multiplied by (the evaluation reference temperature - the data acquisition temperature) / (the second basic temperature - the first basic temperature).

[0222] (Item 11) A condition monitoring device described in any one of items 1 to 7, or item 9, The correction unit is a condition monitoring device that performs linear correction by adding to the current-related data the difference obtained by subtracting the first current-related basic data from the second current-related basic data, multiplied by f(the evaluation reference temperature - the data acquisition temperature) / f(the second basic temperature - the first basic temperature), when the function that represents the current change ΔI in terms of the temperature change Δtp is ΔI = f(Δtp).

[0223] (Item 12) A condition monitoring device described in any one of items 1 to 7, or any one of items 9 to 11, The first current-related basic data, the second current-related basic data, and the current-related data are time-series data of current. The correction unit corrects the current-related data to correspond to the data at a predetermined evaluation reference temperature, corrects the temperature characteristics of the motor's lost torque, or corrects the current corresponding to the motor's lost torque to cancel it out, and is used in this condition monitoring device.

[0224] (Item 13) A condition monitoring device described in any one of items 1 to 7, or any one of items 9 to 12, The correction unit is a condition monitoring device that performs correction on the current-related data by using a relationship in which, when A is a number greater than 1, the current value corresponding to the motor's lost torque becomes 1 / A for every predetermined unit temperature increase in temperature.

[0225] (Item 14) A condition monitoring device described in any one of items 1 to 7, or any one of items 9 to 12, The correction unit is a condition monitoring device that divides the time-series data of the current-related data into a first period in which the rotational speed of the motor is substantially zero and a second period in which the rotational speed of the motor is substantially not zero, and performs different corrections on each of the first and second periods.

[0226] (Item 15) A condition monitoring method for monitoring the state of an industrial robot capable of reproducing predetermined actions, With respect to the current of the motor that drives the industrial robot, first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, are acquired. The first base temperature, which is the temperature at the time of acquiring the first current-related base data, and the second base temperature, which is the temperature at the time of acquiring the second current-related base data, are acquired. With respect to the current of the motor, at least one current data or current-related data which is statistical data of the current data is acquired. The data acquisition temperature, which is the temperature at the time of acquiring the current-related data, is obtained. Based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature, the current-related data is corrected. A condition monitoring method that uses the corrected current-related data for evaluating the condition of a robot.

Claims

1. A condition monitoring device for monitoring the status of an industrial robot capable of reproducing predetermined actions, A current-related basic data acquisition unit acquires, with respect to the current of the motor that drives the industrial robot, first current-related basic data which is at least one current data or statistical data of the current data, and second current-related basic data which is at least one current data or statistical data of the current data. A base temperature acquisition unit that acquires a first base temperature, which is the temperature at the time of acquiring the first current-related base data, and a second base temperature, which is the temperature at the time of acquiring the second current-related base data, A current-related data acquisition unit acquires current-related data, which is at least one current data or statistical data of the current data, for the current of the motor. A temperature acquisition unit that acquires the data acquisition temperature, which is the temperature at the time of acquiring the current-related data, A correction unit that corrects the current-related data to correspond to the data at the evaluation reference temperature, based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature. A robot condition evaluation unit that determines whether or not there is an abnormality in the industrial robot using the corrected current-related data, Equipped with, The robot state evaluation unit, based on the corrected current-related data, The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or The larger of the absolute values ​​of the maximum and minimum current values. A condition monitoring device that calculates a value and determines whether or not there is an abnormality in the industrial robot based on the calculation result.

2. A condition monitoring device for monitoring the status of an industrial robot capable of reproducing predetermined actions, A current-related basic data acquisition unit acquires, with respect to the current of the motor that drives the industrial robot, first current-related basic data which is at least one current data or statistical data of the current data, and second current-related basic data which is at least one current data or statistical data of the current data. A base temperature acquisition unit that acquires a first base temperature, which is the temperature at the time of acquiring the first current-related base data, and a second base temperature, which is the temperature at the time of acquiring the second current-related base data, A current-related data acquisition unit acquires current-related data, which is at least one current data or statistical data of the current data, for the current of the motor. A temperature acquisition unit that acquires the data acquisition temperature, which is the temperature at the time of acquiring the current-related data, A correction unit that corrects the current-related data to correspond to the data at the evaluation reference temperature, based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature. A robot condition evaluation unit that determines whether or not there is an abnormality in the industrial robot using the corrected current-related data, Equipped with, Based on at least one of the first current-related basic data and the second current-related basic data, comparison data to be compared with the current-related data is obtained. The robot state evaluation unit is a state monitoring device that determines the degree of dissimilarity between the current-related data and the comparison target data, and determines whether or not there is an abnormality in the industrial robot based on the degree of dissimilarity.

3. A condition monitoring device according to claim 2, The robot state evaluation unit is a state monitoring device that determines the dissimilarity by taking the Euclidean distance between the current-related data and the comparison target data.

4. A condition monitoring device according to claim 2, The robot state evaluation unit determines the degree of dissimilarity between the current-related data and the comparison data while moving the current-related data relative to the comparison data in multiple steps in the time axis direction. A state monitoring device in which, in a state in which the current-related data has been moved relative to itself so that the dissimilarity is minimized, the robot state evaluation unit determines the dissimilarity between the current-related data and the comparison target data.

5. A condition monitoring device according to claim 2, The robot state evaluation unit is a state monitoring device that determines the DTW distance between the current-related data and the comparison target data as the dissimilarity.

6. A condition monitoring device according to claim 1, A state monitoring device in which the first current-related basic data, the second current-related basic data, and the current-related data are time-series data of current.

7. A condition monitoring device for monitoring the status of an industrial robot capable of reproducing predetermined actions, A current-related basic data acquisition unit acquires, with respect to the current of the motor that drives the industrial robot, first current-related basic data which is at least one current data or statistical data of the current data, and second current-related basic data which is at least one current data or statistical data of the current data. A base temperature acquisition unit that acquires a first base temperature, which is the temperature at the time of acquiring the first current-related base data, and a second base temperature, which is the temperature at the time of acquiring the second current-related base data, A current-related data acquisition unit acquires current-related data, which is at least one current data or statistical data of the current data, for the current of the motor. A temperature acquisition unit that acquires the data acquisition temperature, which is the temperature at the time of acquiring the current-related data, A correction unit that corrects the current-related data to correspond to the data at the evaluation reference temperature, based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature. A robot condition evaluation unit that determines whether or not there is an abnormality in the industrial robot using the corrected current-related data, Equipped with, The first current-related basic data, the second current-related basic data, and the current-related data are The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or A state monitoring device that uses the larger of the absolute values ​​of the maximum and minimum current values.

8. A condition monitoring device according to any one of claims 1 to 7, The correction unit performs a linear correction by adding to the current-related data the difference obtained by subtracting the first current-related basic data from the second current-related basic data, multiplied by (the evaluation reference temperature - the data acquisition temperature) / (the second basic temperature - the first basic temperature).

9. A condition monitoring device according to any one of claims 1 to 6, The correction unit is a condition monitoring device that performs linear correction by adding to the current-related data the difference obtained by subtracting the first current-related basic data from the second current-related basic data, multiplied by f(the evaluation reference temperature - the data acquisition temperature) / f(the second basic temperature - the first basic temperature), when the function that represents the current change ΔI in terms of the temperature change Δtp is ΔI = f(Δtp),

10. A condition monitoring device according to any one of claims 1 to 6, The first current-related basic data, the second current-related basic data, and the current-related data are time-series data of current. The correction unit corrects the current-related data to correspond to the data at a predetermined evaluation reference temperature, corrects the temperature characteristics of the motor's lost torque, or corrects the current corresponding to the motor's lost torque to cancel it out, and is used in this condition monitoring device.

11. A condition monitoring device according to any one of claims 1 to 6, The correction unit is a condition monitoring device that performs correction on the current-related data using a relationship in which, when A is a number greater than 1, the current value corresponding to the motor's lost torque becomes 1 / A for every predetermined unit temperature increase in temperature.

12. A condition monitoring device according to any one of claims 1 to 6, The current-related data mentioned above is time-series data. The correction unit is a condition monitoring device that divides the time-series data of the current-related data into a first period in which the rotational speed of the motor is substantially zero and a second period in which the rotational speed of the motor is substantially not zero, and performs different corrections on each of the first and second periods.

13. A condition monitoring device according to any one of claims 1 to 7, A condition monitoring device that predicts the failure date of the industrial robot based on the determination of whether or not there is an abnormality by the robot condition evaluation unit.

14. A condition monitoring method for monitoring the status of an industrial robot capable of reproducing predetermined actions, With respect to the current of the motor that drives the industrial robot, first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, are acquired. The first base temperature, which is the temperature at the time of acquiring the first current-related base data, and the second base temperature, which is the temperature at the time of acquiring the second current-related base data, are acquired. With respect to the current of the motor, at least one current data or current-related data which is statistical data of the current data is acquired. The data acquisition temperature, which is the temperature at the time of acquiring the current-related data, is obtained. Based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature, the current-related data is corrected to correspond to the data at the evaluation reference temperature. The corrected current-related data is used to determine whether or not there is an abnormality in the industrial robot. Based on the corrected current-related data, The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or The larger of the absolute values ​​of the maximum and minimum current values. A condition monitoring method that calculates a value and determines whether or not there is an abnormality in the industrial robot based on the result of this calculation.

15. A condition monitoring method for monitoring the status of an industrial robot capable of reproducing predetermined actions, With respect to the current of the motor that drives the industrial robot, first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, are acquired. The first base temperature, which is the temperature at the time of acquiring the first current-related base data, and the second base temperature, which is the temperature at the time of acquiring the second current-related base data, are acquired. With respect to the current of the motor, at least one current data or current-related data which is statistical data of the current data is acquired. The data acquisition temperature, which is the temperature at the time of acquiring the current-related data, is obtained. Based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature, the current-related data is corrected to correspond to the data at the evaluation reference temperature. The corrected current-related data is used to determine whether or not there is an abnormality in the industrial robot. Based on at least one of the first current-related basic data and the second current-related basic data, comparison data to be compared with the current-related data is determined. A condition monitoring method that determines the degree of dissimilarity between the current-related data and the comparison target data, and determines whether or not there is an abnormality in the industrial robot based on the degree of dissimilarity.

16. A condition monitoring method for monitoring the status of an industrial robot capable of reproducing predetermined actions, With respect to the current of the motor that drives the industrial robot, first current-related basic data, which is at least one current data or statistical data of the current data, and second current-related basic data, which is at least one current data or statistical data of the current data, are acquired. The first base temperature, which is the temperature at the time of acquiring the first current-related base data, and the second base temperature, which is the temperature at the time of acquiring the second current-related base data, are acquired. With respect to the current of the motor, at least one current data or current-related data which is statistical data of the current data is acquired. The data acquisition temperature, which is the temperature at the time of acquiring the current-related data, is obtained. Based on the first current-related basic data, the second current-related basic data, the first base temperature, the second base temperature, and the data acquisition temperature, the current-related data is corrected to correspond to the data at the evaluation reference temperature. The corrected current-related data is used to determine whether or not there is an abnormality in the industrial robot. The first current-related basic data, the second current-related basic data, and the current-related data are The root mean square of the electric current, Maximum current, Minimum value of current, The value obtained by subtracting the current value at the lower peak from the current value at the higher peak of the current waveform, or A state monitoring method that uses the larger of the absolute values ​​of the maximum and minimum current values.

Images