JP7056418B2 - Abnormality judgment device and abnormality judgment method - Google Patents

Description

The present invention relates to an abnormality determination device and an abnormality determination method for determining an abnormality occurring in a moving part such as a speed reducer provided in a device such as a robot.

Conventionally, as a device for detecting an abnormality generated in a robot, for example, a device disclosed in Patent Document 1 is known. In Patent Document 1, the average value of the disturbance torque generated in the speed reducer mounted on the robot is calculated, and when the calculated average value exceeds a preset threshold value, it is assumed that an abnormality has occurred in the speed reducer. The determination is disclosed.

Japanese Unexamined Patent Publication No. 9-174482

However, the disturbance torque (data on the state of the moving part) generated in the speed reducer may fluctuate greatly during the transitional period after maintenance such as teaching change, equipment repair, and replacement. Therefore, if the threshold value for determining the abnormality is fixed as described in Patent Document 1, there arises a problem that the abnormality generated in the reducer during the transient period cannot be determined with high accuracy.

The present invention has been made to solve such a conventional problem, and an object of the present invention is to obtain high accuracy the abnormality generated in a device even when the data regarding the state of a moving part fluctuates greatly. It is an object of the present invention to provide an abnormality determination apparatus and an abnormality determination method capable of determining an abnormality.

In order to achieve the above object, the present invention identifies a transitional period in which the reference value fluctuates and a stable period in which the reference value stabilizes, based on a reference value indicating a reference value of data regarding the state of the moving part, and after the transitional period. Based on the reference value in the stable period of, the data detected in the transient period is corrected, and the abnormality of the device is determined based on the corrected data.

According to the present invention, even when the data regarding the state of the movable portion fluctuates greatly, it is possible to determine the abnormality occurring in the device with high accuracy.

FIG. 1 is a block diagram showing a configuration of an abnormality determination device according to an embodiment of the present invention and its peripheral devices. FIG. 2 is an explanatory diagram showing an example in which the abnormality determination device shown in FIG. 1 is configured by an integrated computer. FIG. 3 is an explanatory diagram showing a method of calculating the moving average. FIG. 4 is a graph showing fluctuations in disturbance torque when a teaching change is made. FIG. 5 is a graph showing the fluctuation of the disturbance torque when the teaching change is performed, (a) is the first reference waveform in the transient period, (b) is the second reference waveform in the transient period, and (c) is. Shows the corrected disturbance torque during the transition period. FIG. 6 is a flowchart showing a processing procedure of the abnormality determination device according to the embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Explanation of the first embodiment]
FIG. 1 is a block diagram showing a configuration of an abnormality determination device according to an embodiment of the present invention and its peripheral devices. As shown in FIG. 1, the abnormality determination device 102 according to the present embodiment is connected to the robot 101 (device) and the user interface 103 (denoted as “UI” in the figure), and determines the abnormality of the robot 101. , The determination result is displayed on a display (not shown) provided in the user interface 103. Alternatively, an alarm sound is emitted from a speaker (not shown) to notify the user of the occurrence of an abnormality. It should be noted that "determining an abnormality" is a concept that includes not only determining an abnormality that is currently occurring but also predicting an abnormality that will occur in the future.

The robot 101 is, for example, a teaching playback type multi-axis robot. The teaching playback is a function in which an operator actually operates a robot by using a teaching pendant attached to the robot, records and reproduces the operation, and operates the robot. In the present embodiment, a teaching playback type robot will be described as an example, but the present invention is not limited thereto.

As shown in FIG. 1, the robot 101 includes a speed reducer 14 (movable unit), an operation control unit 15, a sensor 13, a disturbance torque calculation unit 12, and a communication unit 11. Although a plurality of speed reducers 14 and sensors 13 are provided in the robot 101, only one is shown in FIG.

The speed reducer 14 includes a servomotor (hereinafter, abbreviated as "motor") for operating the joint axis of the robot arm, and operates under the control of the motion control unit 15. Then, by operating the speed reducer 14, for example, the welding electrode mounted on the tip of the robot arm is brought into contact with a desired portion of an object to be machined (for example, a metal blank material), and welding work is performed. implement. In addition to the welding work, the robot 101 can perform various work such as pressing, painting, resin molding, and assembling the object.

The sensor 13 is installed in the robot 101 and includes, for example, a pulse generator, an encoder, and the like, the position and angle of the robot arm operated by the speed reducer 14, the rotation angle, the rotation speed, and the power consumption of the motor provided in the speed reducer 14. Various physical quantities such as current and rotation angle of the speed reducer 14 are detected. Further, the sensor 13 detects the torque value generated in the motor of the speed reducer 14. The sensor data detected by the sensor 13 is transmitted from the communication unit 11 to the abnormality determination device 102.

The motion control unit 15 operates the speed reducer 14 according to the motion program set by the teaching described above, and controls each robot arm and joint axis mounted on the robot 101 to perform a desired motion. Further, the operation data when the robot 101 is operated is output to the communication unit 11. The operation data includes various information regarding the operation of the robot 101. Details will be described later.

The disturbance torque calculation unit 12 calculates the disturbance torque (data regarding the state of the movable portion) generated in the motor of the speed reducer 14. The disturbance torque indicates the difference between the torque command value when controlling the motor and the torque detection value detected by the sensor 13. When the motor is normal and the speed reducer 14 is operating stably, the difference between the torque command value and the torque detection value is almost constant, so that the disturbance torque shows a stable value. When an abnormality occurs in the speed reducer 14, the speed reducer 14 does not operate stably, and a large change occurs in the disturbance torque. Further, when the maintenance is carried out by the speed reducer 14, a step shift (details will be described later) occurs in the disturbance torque. The disturbance torque is an example of data regarding the state of the movable portion (reducer 14).

The communication unit 11 transmits the operation data of the robot 101, the disturbance torque calculated by the disturbance torque calculation unit 12, and various sensor data detected by the sensor 13 to the abnormality determination device 102.

Each of the above-mentioned functions included in the robot 101 may be implemented by one or more processing circuits. The processing circuit includes a programmed processing device such as a processing device including an electric circuit. Processing devices also include devices such as application specific integrated circuits (ASICs) and conventional circuit components arranged to perform the functions of the robot 101.

The user interface 103 includes an input unit for the operator to perform various operations and a display for displaying various information. For example, when the operator inputs the maintenance data indicating that the maintenance of the robot 101 has been performed in the input unit, the operation of writing the maintenance data to the maintenance history DB 32 described later is performed. As the user interface 103, a tablet terminal can be used.
Note that "maintenance" is a concept that includes changing the teaching of the robot, maintenance such as refueling of grease, and replacement of the speed reducer 14. The teaching change means changing the operation of the speed reducer 14 mounted on the robot 101.

Next, the configuration of the abnormality determination device 102 will be described. The abnormality determination device 102 includes a control unit 51 and various databases (DB). The control unit 51 includes a communication unit 21, an abnormality degree determination unit 22, a moving average calculation unit 23, a steady-time average calculation unit 24, a transient period detection unit 25, an amplitude component extraction unit 26, and a disturbance torque correction unit. 27 and an alarm determination unit 28 are included. The database includes a sensor DB 31, a maintenance history DB 32, and an operation history DB 33.

The sensor DB 31 stores sensor data such as the position and angle of the robot arm detected by the sensor 13, the rotation angle of the motor, and the rotation speed. Further, the disturbance torque calculated by the disturbance torque calculation unit 12 is stored.

The operation history DB 33 stores the operation data of the robot 101. The operation data includes various data related to the operation such as the operation date of the robot 101, the time when the operation is started, the time when the operation is stopped, the time when the robot is continuously operated, and the time when the robot 101 is continuously stopped. Further, the operation data includes the operation mode of the speed reducer 14. The operation mode includes a normal operation mode, a maintenance mode, and a stop mode.

The maintenance history DB 32 stores maintenance data when maintenance is performed on the robot 101 when an abnormality occurs in the speed reducer 14 or when an abnormality is predicted to occur. As described above, the maintenance data includes the teaching change of the robot 101. The maintenance data can be input by the operator through the input unit (not shown) of the user interface 103. Alternatively, when the robot 101 is operated in the maintenance mode described above, it may be determined that maintenance has been performed, and maintenance data may be automatically created and stored.

The maintenance data includes the ID number of the speed reducer 14 that has been maintained, the ID number of the motor mounted on the speed reducer 14, the date and time when the maintenance was performed, and the content of the maintenance (replacement, repair, grease refueling, instruction). Changes etc.) are included. Further, the maintenance data includes data related to the teaching change when the speed reducer 14 is taught and changed. For example, data of the operation amount of the speed reducer 14 before the instruction change and the operation amount of the speed reducer 14 after the instruction change are included.

The communication unit 21 communicates with the communication unit 11 provided in the robot 101. The communication unit 21 receives the operation data of the robot 101 transmitted from the robot 101 and outputs it to the operation history DB 33. Further, the communication unit 21 receives the disturbance torque and the sensor data transmitted from the robot 101 and outputs them to the sensor DB 31.

The abnormality degree determination unit 22 acquires the past disturbance torque of the motor mounted on the speed reducer 14 from the sensor DB 31, and calculates the abnormality degree indicating the abnormality of the acquired disturbance torque.
Hereinafter, the method of calculating the degree of abnormality will be described. The degree of abnormality a (x') when the disturbance torque is x'is defined by the following equation (1).
a (x') = {(x'-m) ² } / 2 ・ s ² ... (1)
However, m is the sample average of the disturbance torque, and s is the standard deviation of the disturbance torque.
Then, when the degree of abnormality a (x') exceeds a certain value, it is determined that this disturbance torque is abnormal.

In addition to the above, kernel density estimation, density ratio estimation, and the like can be used as a method for calculating the degree of abnormality. Further, as another method for calculating the degree of abnormality, the difference between the disturbance torque and the predetermined torque threshold value is calculated, and further, the rate of change of this difference with respect to the passage of time is calculated. Then, when the calculated rate of change exceeds a predetermined rate of change threshold, it can be determined that the disturbance torque is abnormal. For the above torque threshold value, for example, the average value of the disturbance torque of the same month one year ago can be used.

Further, when the abnormality degree determination unit 22 determines the abnormality of the disturbance torque, the maintenance history DB 32 is referred to, and the cause of the abnormality is due to maintenance based on the maintenance data performed before the abnormality determination. It is determined whether the data is due to non-maintenance or not, and the data indicating the abnormality is labeled and stored in the operation history DB 33. For example, if maintenance has been carried out on the speed reducer 14 within the past one week from the abnormality determination, a label indicating "maintenance carried out" is attached to the data indicating the abnormality. On the other hand, if maintenance has not been carried out within the past week, it will be labeled as "maintenance not carried out".
Further, the abnormality degree determination unit 22 calculates the abnormality degree by the same method as described above based on the disturbance torque during the transient period corrected by the disturbance torque correction unit 27 described later, and determines the occurrence of the abnormality.

The moving average calculation unit 23 calculates the moving average of the disturbance torque detected by the speed reducer 14. The "moving average" is the average of the disturbance torque over a certain period of time. FIG. 3 is an explanatory diagram showing the moving average of the disturbance torque Q1. When the current time is t1, the average Av1 of the disturbance torque of the period Tx1 retroactive from the time t1 by a predetermined period is calculated as the moving average. To. That is, the moving average at time t1 is Av1. At the subsequent time t2, the average Av2 of the disturbance torque of the period Tx2 that goes back by a predetermined period from this time t2 is calculated as a moving average. Similarly, at time t3, the average Av3 of the disturbance torque of the period Tx3 is calculated as a moving average. In this way, the moving average shows the average value to be calculated by discarding the oldest data every time new data is obtained over time, and is updated to a new numerical value with the passage of time.

Based on the operation data of the robot 101 stored in the operation history DB 33, the constant-time average calculation unit 24 changes the teaching by the speed reducer 14, and after that, it is a fixed period in a steady state in which a long time has passed. Calculate the average value of the disturbance torque during the stable period. The calculated average value is taken as the steady-state average. Specifically, the average value of the disturbance torque after the time t12 (the time when the transient period ends) shown in FIG. 4 described later is defined as the steady-state average.

The transition period detection unit 25 determines the disturbance torque after maintenance is performed based on the maintenance data performed by the speed reducer 14 stored in the maintenance history DB 32 and the disturbance torque generated in the speed reducer 14. Detects a transient period, which is the period during which the average value fluctuates. The average value of the disturbance torque is an example of a reference value indicating a reference of "data on the state of moving parts". In addition to the average value, the median value of the amplitude of the disturbance torque (the value at the center of the maximum amplitude and the minimum amplitude) may be used as a reference value.

As a specific example of detecting the transient period, the moving average calculated by the moving average calculation unit 23 after the teaching change is acquired, and the transition period is set when the slope of the moving average with respect to the passage of time is larger than the first threshold value. Judge that it exists. For example, when the teaching change is carried out at the time t11 shown in FIG. 4, a transitional period exists when the slope d1 (negative slope in this example) of the moving average thereafter is larger than the first threshold value. Judge.
Alternatively, when the difference between the moving average and the above-mentioned steady-time average is larger than the second threshold value, it is determined that a transient period exists. For example, when the difference g1 between the moving average and the steady time average P3 at time t11 shown in FIG. 4 is larger than the second threshold value, it is determined that a transient period exists.

As a method for calculating the transient period, the steady-time average calculated by the steady-time average calculation unit 24 is acquired, the difference between the moving average of the disturbance torque and the steady-time average after the teaching change is calculated, and this difference is preset. The period until the value becomes equal to or less than the third threshold value (zero or a numerical value close to zero) is detected as a transient period. Hereinafter, a detailed description will be given with reference to FIG.

FIG. 4 is a waveform diagram showing changes in the disturbance torque generated in the speed reducer 14 with respect to the passage of time, and shows a case where the teaching change (maintenance) of the robot 101 is performed at time t11. As shown in FIG. 4, when the teaching change of the robot 101 is carried out at time t11, a large change (hereinafter referred to as “step deviation”) occurs in the disturbance torque. Further, after the teaching change is carried out at time t11, the disturbance torque overshoots, and then gradually decreases and converges to the steady disturbance torque Tg. Further, when the slope d1 of the moving average of the disturbance torque after overshoot is larger than the above-mentioned first threshold value, it is determined that there is a transient period. Further, at time t12, the disturbance torque substantially coincides with the steady-state average and becomes equal to or less than the above-mentioned third threshold value. Therefore, the time t11 to the time t12 is set as the transient period Tc.

Further, as another method for setting the transition period, it is also possible to set the period until the rate of change (slope) of the moving average with respect to the passage of time becomes equal to or less than a predetermined value as the transition period. Further, it is also possible to set the transient period according to the operating amount of the speed reducer 14 by changing the teaching. Specifically, the difference between the operating amount of the speed reducer 14 before the teaching change is carried out and the operating amount of the speed reducer 14 after the teaching change is carried out is calculated, and the larger the difference, the longer the transient period is set. You may.

The amplitude component extraction unit 26 removes spikes (sudden fluctuation components) from the disturbance torque, and further calculates the difference between the moving average calculated by the moving average calculation unit 23 and the disturbance torque after removing the spikes, and disturbs the disturbance. Extract the torque amplitude component. The "amplitude component" refers to a component that is displaced in the vertical direction with respect to a moving average. Further, the "spike" is a noise component of the disturbance torque that is suddenly generated regardless of the abnormality of the speed reducer 14.

For example, when the speed reducer 14 is stopped for a long time and then started to operate, spikes occur for a long period of time from the start of operation, and the frequency of spikes increases. Similarly, when the magnitude of the load driven by the speed reducer 14 is changed or when the temperature of the speed reducer 14 is low, the frequency of spike occurrence increases. When it is predicted that the frequency of spikes will increase, it is determined that many spikes included in the disturbance torque are superimposed, and a calculation for removing the spikes is performed. By removing the spikes, the amplitude component from which the noise component has been removed can be extracted.

As the temperature of the speed reducer 14, the temperature in the factory where the speed reducer 14 is installed or the temperature in the area obtained via the network or the like can be used. In addition, a map showing the temperature of the reducer 14 and the relationship between the operation data of the reducer 14 and the spike is prepared in advance, and when the temperature of the reducer 14 and the operation data are acquired, these data are used as a map. Spikes can also be removed by applying.

In addition, as another method for removing spikes, a moving average of the disturbance torque is calculated after the teaching change is performed, and the disturbance torque that changes by a predetermined ratio or more (for example, ± 20% or more) with respect to the moving average. If a waveform exists, it can be judged as a spike and replaced with a moving average value to remove the spike.
That is, the conditions for detecting spikes are changed according to at least one of the operating state of the speed reducer 14, the temperature of the speed reducer 14, and the presence or absence of maintenance for the speed reducer 14.

The disturbance torque correction unit 27 corrects the disturbance torque in the transient period Tc detected by the transient period detection unit 25. In the above-mentioned transient period Tc, the moving average waveform calculated by the moving average calculation unit 23 (this is referred to as “first reference waveform P1”) is set. As a result, as shown in FIG. 5A, the first reference waveform P1 in which the above-mentioned amplitude component is removed from the disturbance torque in the transient period Tc is generated.

Further, the disturbance torque correction unit 27 shifts the first reference waveform P1 so that the first reference waveform P1 in the transient period Tc coincides with the steady-state average P3. Specifically, as shown in FIG. 5B, the first reference waveform P1 is shifted so as to match the steady-state average P3. The waveform after the shift is referred to as the second reference waveform P2.

After that, the disturbance torque correction unit 27 superimposes the amplitude component extracted by the amplitude component extraction unit 26 on the second reference waveform P2 in the transient period Tc, and generates the disturbance torque in the transient period Tc. As a result, as shown in FIG. 5C, a disturbance torque having a waveform whose moving average substantially coincides with the steady-state average P3 can be obtained in the transient period Tc.

The alarm determination unit 28 determines whether or not to output an alarm for notifying the abnormality when the abnormality degree determination unit 22 determines that the abnormality is abnormal based on the disturbance torque corrected by the disturbance torque correction unit 27. If it is determined and it is determined to output an alarm, an alarm output command is output to the user interface 103.

Here, the abnormality determination device 102 may be realized by using a computer including a CPU 41 (central processing unit), a memory 42, and each database (sensor DB 31, maintenance history DB 32, operation history DB 33) as shown in FIG. can. A computer program (abnormality determination program) for making the computer function as the abnormality determination device 102 is installed and executed in the computer. As a result, the CPU 41 has a plurality of information processing circuits included in the abnormality determination device 102, that is, a communication unit 21, an abnormality degree determination unit 22, a moving average calculation unit 23, a steady-time average calculation unit 24, a transient period detection unit 25, and a disturbance. It functions as a torque correction unit 27 and an alarm determination unit 28.

Each of the above-mentioned functions included in the abnormality determination device 102 can be implemented by one or a plurality of processing circuits. The processing circuit includes a programmed processing device such as a processing device including an electric circuit. Processing devices also include devices such as application specific integrated circuits (ASICs) and conventional circuit components arranged to perform the functions of the anomaly determination device 102.

[Explanation of operation of the first embodiment]
Next, the processing procedure of the abnormality determination device according to the present embodiment will be described with reference to the flowchart shown in FIG. An example in which the process shown in FIG. 6 is executed after a while has passed since the abnormality occurred in the speed reducer 14 will be described. However, it can also be executed in real time.

First, in step S11, the abnormality degree determination unit 22 calculates the abnormality degree of the speed reducer 14 based on the disturbance torque of the speed reducer 14 stored in the sensor DB 31. The degree of abnormality can be calculated using the above-mentioned equation (1).

In step S12, the abnormality degree determination unit 22 determines whether or not an abnormality has occurred in the speed reducer 14. When the degree of abnormality calculated in the process of step S11 exceeds a certain value, it is determined that an abnormality has occurred in the speed reducer 14 (YES in step S12), and the process proceeds to step S13.

In step S13, the abnormality degree determination unit 22 acquires the data indicating the abnormality stored in the operation history DB 33. As described above, the operation history DB 33 stores data indicating an abnormality, and is further labeled with "maintenance carried out" and "maintenance not carried out". Based on this label, it is recognized whether the abnormality generated in the speed reducer 14 is an abnormality caused by performing maintenance or an abnormality caused by a cause other than maintenance.

In step S14, it is determined whether or not the label "maintenance is carried out" is attached. When the label of "maintenance is carried out", the maintenance of the speed reducer 14 (for example, change of teaching) is carried out within a predetermined period (for example, one week) from the time when the occurrence of the abnormality is detected. It is recognized that it has been done.

If maintenance has not been performed (NO in step S14), it is determined that the cause of the abnormality detected is not maintenance, and the process proceeds to other factor analysis processing in step S15.

When the maintenance is carried out (YES in step S14), the moving average calculation unit 23 calculates the moving average after the maintenance is carried out in step S16. That is, when maintenance is carried out, as shown in FIG. 4, there is a high possibility that the disturbance torque has a step shift due to the maintenance, so is there a transitional period after the step shift occurs? A moving average is calculated to determine whether or not. For example, the moving average is calculated in a certain period after the step deviation occurs. The method of calculating the moving average is as described with reference to FIG. By this calculation, as shown in FIG. 4, the moving average from the time t11 when the step shift occurs to the lapse of a certain period is obtained.

In step S17, the transient period detection unit 25 determines whether or not there is a transient period after the maintenance is performed, based on the moving average calculated in the process of step S16. Specifically, as described above, the slope of the moving average of the disturbance torque after the maintenance is performed (d1 shown in FIG. 4), or the difference between the moving average during the transient period and the steady average (g1 shown in FIG. 4). It is determined whether or not there is a transition period based on. If it is determined that the transient period exists, the process proceeds to step S18, and if it is not determined that the transient period exists, the process proceeds to step S22.

In step S18, the transient period detection unit 25 sets the length of the transient period. As described above, the period until the difference between the moving average and the steady-time average reaches the third threshold value (zero or a numerical value close to zero) is set as the transient period. Specifically, as shown in FIG. 4, a period from time t11 when maintenance is performed and a step shift occurs in the disturbance torque Tg to time t12 when the moving average of the disturbance torque Tg almost coincides with the steady-state average P3. Is set as the transition period Tc. Further, as described above, the transient period may be set according to the change in the operating amount of the speed reducer 14 due to the change in teaching.

In step S19, the amplitude component extraction unit 26 extracts the amplitude component of the disturbance torque in the transient period Tc, and further removes the spike included in the disturbance torque. Of the amplitude components included in the disturbance torque, sudden fluctuation components are regarded as spikes and removed. As described above, the amplitude component extraction unit 26 refers to the operation history of the robot 101 from the operation history DB 33, refers to the stop time, the operation time, and the load state of the speed reducer 14, and removes the spikes. Alternatively, the temperature data of the speed reducer 14 is acquired, and the spikes are removed by referring to the temperature of the speed reducer 14 or the ambient temperature of the speed reducer 14.

In step S20, the disturbance torque correction unit 27 generates the first reference waveform P1 shown in FIG. 5A based on the moving average in the transient period Tc, and converts the first reference waveform P1 into the steady-state average P3. shift. As a result, the second reference waveform P2 shown in FIG. 5B is obtained.

In step S21, the disturbance torque correction unit 27 superimposes the amplitude component of the disturbance torque on the second reference waveform P2. As a result, as shown in FIG. 5C, a disturbance torque whose moving average coincides with the steady-state average P3 is obtained in the transient period Tc.

In step S22, the abnormality degree determination unit 22 determines an abnormality based on the disturbance torque generated in the process of step S21. If it is determined that there is no abnormality, the process returns to step S11, and if it is determined that there is an abnormality, the process proceeds to step S23.

In step S23, the alarm determination unit 28 determines whether or not to output an alarm, and if it is determined to output an alarm, outputs an alarm signal in step S24. As a result, the occurrence of an abnormality is notified to the display or the like provided in the user interface 103 shown in FIG.

In step S25, the abnormality degree determination unit 22 stores in the operation history DB 33 that an abnormality has occurred in the speed reducer 14. In this way, it is possible to detect an abnormality occurring in the speed reducer 14 and notify the user.

In the above-described embodiment, an example of performing the process of setting the transition period in step S18 of FIG. 6 has been described, but when the transition period is set to a certain period, the process of step S18 becomes unnecessary. ..

In this way, the abnormality determination device according to the present embodiment can obtain the following effects.
(1)
In the abnormality determination device 102 according to the present embodiment, when the average value of the disturbance torque of the speed reducer 14 (reference value of data regarding the state of the movable part) fluctuates, that is, the disturbance torque is changed as shown at time t11 in FIG. When a step shift occurs, the period from the occurrence of the step shift until the average value of the disturbance torque becomes a steady state (transient period) and the period after the reference value becomes a steady state (stable). Period). Then, the disturbance torque in the transient period is corrected based on the reference value (for example, the average value) of the disturbance torque in the stable period. The abnormality of the speed reducer 14 is determined based on the corrected disturbance torque. Therefore, even if a step shift occurs in the disturbance torque due to the implementation of the teaching change or the like, it is possible to detect the abnormality of the speed reducer 14 with high accuracy.
In particular, maintenance data is not real-time and may be input after the actual maintenance work is completed, so the latest maintenance data can be used by making an abnormality judgment after a lapse of time, and more accurate abnormalities can be used. It becomes possible to make a judgment.

(2)
By setting the reference value of the disturbance torque as the average value or the median of the amplitude, the reference value can be approximately the center of the amplitude of the disturbance torque that fluctuates up and down, and highly accurate abnormality detection becomes possible.

(3)
Since the moving average of the disturbance torque in the transition period is calculated and the disturbance torque in the transition period is corrected so that the calculated moving average matches the average value in the stable period, the fluctuation amount of the disturbance torque in the transition period can be offset. More accurate abnormality detection is possible.

(4)
During the transition period, spikes (sudden fluctuation components) contained in the disturbance torque are removed, and the moving average is calculated based on the disturbance torque after the spikes are removed, so the influence of spikes can be avoided and the moving average can be made highly accurate. Can be calculated. As a result, highly accurate abnormality detection becomes possible.

(5)
Since the conditions for detecting spikes are changed according to the operating state of the speed reducer 14, the temperature of the speed reducer 14, or the presence or absence of maintenance in the speed reducer 14, spike removal suitable for the situation of the speed reducer 14 becomes possible.

(6)
As the operating state of the speed reducer 14, the conditions for detecting spikes are changed according to the stop time, the operating time, and the magnitude of the load of the speed reducer 14. Specifically, when the speed reducer 14 is stopped for a long time and then started to operate, spikes are generated for a long period from the start of operation, and the frequency of spikes is increased. Similarly, when the magnitude of the load driven by the speed reducer 14 changes, the frequency of spike occurrence increases. Therefore, since the spikes are detected on the premise that many spikes are present in the disturbance torque, the spike removal accuracy can be improved.

(7)
Since the ambient temperature of the speed reducer 14 is used as the temperature of the speed reducer 14, the temperature of the speed reducer 14 can be obtained by a simple method. Further, since the frequency of spike generation increases as the temperature of the speed reducer 14 decreases, the spike removal accuracy can be improved by removing the spikes according to the temperature of the speed reducer 14.

(8)
When the disturbance torque of the speed reducer 14 is deviated due to the implementation of maintenance such as changing the teaching, the rate of change of the moving average of the disturbance torque (slope d1 shown in FIG. 4) after the maintenance is performed is calculated. Then, the period until the rate of change becomes equal to or less than the predetermined value is set as the transition period. Specifically, the time t11 to t12 shown in FIG. 4 is set as the transient period Tc. Therefore, the transient period can be set with high accuracy. Therefore, it is possible to perform abnormality detection in the transient period with high accuracy.

In the above-described embodiment, the case where the disturbance torque step deviation (variation of the reference value of the disturbance torque) occurs after the maintenance such as the instruction change is performed has been described as an example, but the present invention has been described as an example. It is not limited to the step deviation that occurs after the implementation of the above, and it can also be applied to the case where the step deviation occurs in the disturbance torque due to other causes.

Further, in the above-described embodiment, as shown in FIG. 4, an example in which the disturbance torque is shifted in the direction of increase (upward in the figure) due to the implementation of the teaching change has been described, but the present invention is limited to this. It can be applied even when the step is shifted in the descending direction (downward in the figure).

[Explanation of modified example]
Next, a modified example of the abnormality determination device according to the above-described embodiment will be described. In the modified example, when the teaching change is performed for the speed reducer 14, the difference between the operating amount of the speed reducer 14 before the teaching change and the operating amount of the speed reducer 14 after the teaching change is calculated, and the calculated difference is obtained. And set the length of the transition period. Specifically, the larger the difference, the longer the transient period is set.

That is, when the teaching change is made to increase the operating amount of the speed reducer 14 such as the rotation angle and the operating region, the amount of change in the disturbance torque becomes large and it takes a long time to reach a steady state. Therefore, a long transient period is set.
Then, by correcting the disturbance torque in the set transient period, more accurate abnormality detection becomes possible.

Here, in the above-described embodiment, an example of determining an abnormality of the robot 101 has been described, but the device for determining the abnormality is not limited to the robot 101. For example, an automobile engine may be used instead of the motor, and a transmission may be used instead of the speed reducer 14. Further, a rotating mechanism of a moving body, a moving body such as a playset in an amusement park, a machine tool such as a three-dimensional printer, that is, all devices having a rotating mechanism and a mechanism for transmitting the rotation mechanism can be targeted. It may also target other types of equipment.

Further, the abnormality determination device may be arranged at a remote location, and necessary signals and data may be transmitted and received via a communication line to determine the abnormality of the device. Further, the abnormality of a plurality of devices may be diagnosed by one abnormality determination device. Further, the plurality of devices may be arranged in different places from each other.

Although embodiments of the present invention have been described above, the statements and drawings that form part of this disclosure should not be understood to limit the invention. This disclosure will reveal to those skilled in the art various alternative embodiments, examples and operational techniques.

11 Communication unit 12 Disturbance torque calculation unit 13 Sensor 14 Reducer (moving part)
15 Operation control unit 21 Communication unit 22 Abnormality determination unit 23 Moving average calculation unit 24 Constant constant average calculation unit 25 Transient period detection unit 26 Amplitude component extraction unit 27 Disturbance torque correction unit 28 Alarm judgment unit 31 Sensor DB
32 Maintenance history DB
33 Operation history DB
51 Control unit 101 Robot (equipment)
102 Abnormality determination device 103 User interface

Claims (10)

An abnormality determination device including a control unit for determining an abnormality of the device based on data related to the state of the movable portion acquired from a sensor installed in the device having the movable portion.
The control unit
Based on the reference value indicating the reference value of the data, the transient period in which the reference value fluctuates and the stable period in which the reference value stabilizes are identified.
The data in the transient period is corrected based on the reference value in the stable period after the transient period.
An abnormality determination device for determining an abnormality of the device based on the corrected data. The abnormality determination device according to claim 1, wherein the reference value is an average value of the data or a median value of the amplitude of the data. The reference value is an average value of the data, and is
The control unit
The moving average of the data is calculated as the average value of the data in the transient period.
The abnormality determination device according to claim 2, wherein the data is corrected so that the moving average matches the average value in the stable period. The control unit
Sudden fluctuation components contained in the data during the transient period are detected.
The abnormality determination device according to claim 3, further comprising calculating the moving average after removing the detected sudden fluctuation component from the data. The control unit may change the conditions for detecting the sudden fluctuation component according to at least one of the operating state of the movable portion, the temperature of the movable portion, and the presence or absence of maintenance of the movable portion. The abnormality determination device according to claim 4. The fifth aspect of the present invention is characterized in that the operating state of the movable portion is at least one of the stop time of the movable portion, the operating time of the movable portion, and the magnitude of the load applied to the movable portion. The above-mentioned abnormality determination device. The abnormality determination device according to claim 5, wherein the temperature of the movable portion is the temperature around the movable portion. The control unit calculates the rate of change of the moving average with respect to the passage of time, and sets the period from the occurrence of the fluctuation of the moving average to the decrease of the rate of change to a predetermined value or less as a transient period. The abnormality determination device according to any one of claims 3 to 7. Maintenance for the movable part is a teaching change that changes the operation of the movable part.
The claim is characterized in that the control unit sets the transient period according to the difference between the amount of movement of the movable part before the change of teaching and the amount of movement of the movable part after the change of teaching. The abnormality determination device according to 5 . It is an abnormality determination method for determining an abnormality of the device based on data related to the state of the movable part acquired from a sensor installed in the device having the movable part.
Based on the reference value indicating the reference value of the data, the transient period in which the reference value fluctuates and the stable period in which the reference value stabilizes are identified.
An abnormality determination method, characterized in that the data in the transition period is corrected based on the reference value in the stable period after the transition period, and the abnormality of the device is determined based on the corrected data.

Images