Robot joint current anomaly detection method and device based on statistical process control

Abstract

The present invention relates to a method and device for detecting abnormal current in robot joints based on statistical process control. The method comprises the following steps: acquiring current signals from industrial robot joints; obtaining the period of the industrial robot joint current signals using autocorrelation analysis; calculating and eliminating the phase difference between the industrial robot joint current monitoring signals and a reference signal using cross-correlation analysis; calculating the total energy of the industrial robot joint current monitoring signals and the total energy of the industrial robot joint current reference signal within the period based on the acquired current signal period, and performing abnormality detection using statistical process control. Compared with existing technologies, the present invention has the advantages of being easy to control, facilitating the generation of time series data, enabling long-term tracking, and providing accurate detection.

Description

Technical Field

[0001] The present invention relates to the field of industrial robot status monitoring, and in particular to a method and device for detecting abnormal current in robot joints based on statistical process control. Background Art

[0002] With the continuous advancement of automation, industrial robots are playing an increasingly important role, significantly improving production efficiency. However, the functional requirements for industrial robots are constantly increasing, their structures are becoming increasingly complex, and long-term operation can lead to industrial robot failures. Condition monitoring of industrial robots is necessary to detect equipment anomalies and prevent economic losses and casualties. Traditional industrial robot anomaly detection methods collect signals such as vibration, torque, and acoustic emission signals. A Chinese patent application (publication number CN 111975784 A) discloses a joint robot fault diagnosis method based on current and vibration signals. This method uses collected joint motor signals to generate a uniform sampling time sequence and then samples the filtered vibration signal at uniform angles. Finally, the vibration signal is Fourier transformed to obtain the order spectrum of the robot joint vibration signal and analyze it to perform joint fault diagnosis. However, the quality of the collected signals is difficult to guarantee due to factors such as the acquisition location. In addition, the sensor cost is relatively high. However, current signals can be directly acquired from the industrial robot control cabinet, which is convenient and low-cost.

[0003] There is considerable research on fault diagnosis and condition monitoring of constant-speed motors using current signals. After acquiring the stator current, current spectrum analysis can be used to perform fault diagnosis and condition monitoring of the motor. However, the joints of industrial robots are primarily composed of servo motors and reducers, and relatively little research has been conducted on condition monitoring and fault diagnosis of reducers and servo motors based on current signals. A Chinese patent application (publication number CN 108638128 A) discloses a real-time anomaly monitoring method and system for industrial robots. This method uses collected joint current signals to calculate their positioning deviation, current boundary, range, and variance, and then compares them with their normal range to complete anomaly detection. However, this method can only process relatively stable signals. Compared to the current signals of servo motors and constant-speed motors, their current signals have periodicity with the action cycle, complex frequency components, and drastic amplitude variations. Using conventional methods to perform anomaly detection based on this current information can result in inaccurate detection results. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method and device for detecting abnormal current in robot joints based on statistical process control.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for detecting abnormal current in robot joints based on statistical process control, characterized by comprising the following steps:

[0007] S1, obtain the current signal of the industrial robot joint;

[0008] S2. Obtain the period of the current signal of the industrial robot joint using the autocorrelation analysis method according to the current signal of the industrial robot joint;

[0009] S3. Calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal using a cross-correlation analysis method based on the industrial robot joint current signal;

[0010] S4. Calculate the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the period based on the period obtained in step S2 and the current signal with the phase difference eliminated in step S3;

[0011] S5. Calculate the control threshold range using the statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal, and make an abnormality judgment based on the control threshold range.

[0012] Furthermore, the specific steps of the statistical process control method used in step S5 are as follows:

[0013] A1. Based on the total energy of the industrial robot joint current reference signal, calculate the upper and lower control limits of the control chart. The calculation formula is as follows:

[0014]

[0015]

[0016]

[0017]

[0018] Among them, t i represents the energy of the i-th reference signal, represents the average value of the total energy within the period of the industrial robot's normal joint current reference signal, σ represents the standard deviation of the total energy within the period of the industrial robot's normal joint current reference signal, UCL represents the upper control limit, and LCL represents the lower control limit;

[0019] A2. Determine whether the total energy of the joint current monitoring signal of the industrial robot is between the upper control limit and the lower control limit. If so, determine that there is no abnormality in the industrial robot; if not, determine that there is an abnormality in the industrial robot.

[0020] Furthermore, in step S1, the method for obtaining the current signal is to pass the U-phase and V-phase cables in the industrial robot control cabinet through the current transformer to collect the U-phase and V-phase currents of each joint.

[0021] Furthermore, the specific steps of the autocorrelation analysis method in step S2 are to obtain the autocorrelation function of the industrial robot joint current signal, then normalize the sequence length to eliminate the influence of time lag, and finally obtain the index distance between the maximum peaks of the autocorrelation function of the industrial robot joint current signal to obtain the number of data points of one cycle of the industrial robot joint current signal and the time of one cycle of the industrial robot joint current signal;

[0022] The calculation expression of the autocorrelation function is as follows:

[0023]

[0024] Where h is the order, μ is the mean of the sequence, and x is the time series of the input signal.

[0025] Furthermore, the cross-correlation function of the cross-correlation analysis method in step S3 is The calculation expression is as follows:

[0026]

[0027] Where f(x) represents the industrial robot joint current monitoring signal, g(x) represents the industrial robot joint current reference signal in normal state, and m represents the phase difference between the signals.

[0028] A robot joint current anomaly detection device based on statistical process control includes a memory and a processor; the memory is used to store a computer program; the processor is used to implement the following method when executing the computer program:

[0029] S1, obtain the current signal of the industrial robot joint;

[0030] S2. Obtain the period of the current signal of the industrial robot joint using the autocorrelation analysis method according to the current signal of the industrial robot joint;

[0031] S3. Calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal using a cross-correlation analysis method based on the industrial robot joint current signal;

[0032] S4. Calculate the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the period based on the period obtained in step S2 and the current signal with the phase difference eliminated in step S3;

[0033] S5. Calculate the control threshold range using the statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal, and make an abnormality judgment based on the control threshold range.

[0034] Furthermore, the specific steps of the statistical process control method used in step S5 are as follows:

[0035] A1. Based on the total energy of the industrial robot joint current reference signal, calculate the upper and lower control limits of the control chart. The calculation formula is as follows:

[0036]

[0037]

[0038]

[0039]

[0040] Among them, t i represents the energy of the i-th reference signal, represents the average value of the total energy within the period of the industrial robot's normal joint current reference signal, σ represents the standard deviation of the total energy within the period of the industrial robot's normal joint current reference signal, UCL represents the upper control limit, and LCL represents the lower control limit;

[0041] A2. Determine whether the total energy of the joint current monitoring signal of the industrial robot is between the upper control limit and the lower control limit. If so, determine that there is no abnormality in the industrial robot; if not, determine that there is an abnormality in the industrial robot.

[0042] Furthermore, in step S1, the method for obtaining the current signal is to pass the U-phase and V-phase cables in the industrial robot control cabinet through the current transformer to collect the U-phase and V-phase currents of each joint.

[0043] Furthermore, the specific steps of the autocorrelation analysis method in step S2 are to obtain the autocorrelation function of the industrial robot joint current signal, then normalize the sequence length to eliminate the influence of time lag, and finally obtain the index distance between the maximum peaks of the autocorrelation function of the industrial robot joint current signal to obtain the number of data points of one cycle of the industrial robot joint current signal and the time of one cycle of the industrial robot joint current signal;

[0044] The calculation expression of the autocorrelation function is as follows:

[0045]

[0046] Where h is the order, μ is the mean of the sequence, and x is the time series of the input signal.

[0047] Furthermore, the cross-correlation function of the cross-correlation analysis method in step S3 is The calculation expression is as follows:

[0048]

[0049] Where f(x) represents the industrial robot joint current monitoring signal, g(x) represents the industrial robot joint current reference signal in normal state, and m represents the phase difference between the signals.

[0050] Compared with the prior art, the present invention has the following advantages:

[0051] The present invention successively obtains the period of the current signal through the autocorrelation analysis method and the cross-correlation analysis method, eliminates the phase difference between the reference signal and the monitoring signal, and finally determines the threshold value of abnormal judgment through the statistical process control method, processes the complex frequency components in the current signal of the complex servo motor, and avoids the interference caused by the drastic amplitude change through the statistical process control method, thereby improving the accuracy of abnormality detection. Moreover, the abnormality detection is completed by relying solely on the collection of the current signal, with low cost, small calculation amount, and good economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of the process of the present invention.

[0053] Figure 2 This is the time domain diagram of the joint current reference signal 1_U of the industrial robot in normal state.

[0054] Figure 3 This is the time domain diagram of the joint current reference signal 2_U of the industrial robot in normal state.

[0055] Figure 4 This is the time domain diagram of the joint current reference signal 3_U of the industrial robot in normal state.

[0056] Figure 5 This is the time domain diagram of the joint current reference signal 4_U of the industrial robot in normal state.

[0057] Figure 6 This is the time domain diagram of the joint current reference signal 5_U of the industrial robot in normal state.

[0058] Figure 7 This is the time domain diagram of the joint current reference signal 6_U of the industrial robot in normal state.

[0059] Figure 8 This is the time domain diagram of the joint current reference signal 1_V of the industrial robot in normal state.

[0060] Figure 9This is the time domain diagram of the joint current reference signal 2_V of the industrial robot in normal state.

[0061] Figure 10 This is the time domain diagram of the 3_V joint current reference signal of the industrial robot in normal state.

[0062] Figure 11 This is the time domain diagram of the 4_V joint current reference signal of the industrial robot in normal state.

[0063] Figure 12 This is the time domain diagram of the 5_V joint current reference signal of the industrial robot in normal state.

[0064] Figure 13 This is the time domain diagram of the joint current reference signal 6_V of the industrial robot in normal state.

[0065] Figure 14 This is the autocorrelation function image of the joint current reference signal 1_U of the industrial robot in normal state.

[0066] Figure 15 This is the segmentation image of the joint current reference signal 1_U of the industrial robot in normal state.

[0067] Figure 16 This is a time domain image comparison diagram of the industrial robot's normal state current reference signal 1_U and the fault state joint current monitoring signal 1_U.

[0068] Figure 17 This is a time domain image comparison diagram of the industrial robot's normal state current reference signal 1_U and the fault state joint current monitoring signal 1_U after eliminating the phase difference.

[0069] Figure 18 This is the control diagram of the industrial robot current signal 1_U.

[0070] Figure 19 This is the control diagram of the industrial robot current signal 1_V.

[0071] Figure 20 This is the control diagram of the industrial robot current signal 2_U.

[0072] Figure 21 This is the control diagram of the industrial robot current signal 2_V.

[0073] Figure 22 This is the control diagram of the industrial robot current signal 3_U.

[0074] Figure 23 This is the control diagram of the industrial robot current signal 3_V.

[0075] Figure 24 This is the control diagram of the industrial robot current signal 4_U.

[0076] Figure 25 This is the control diagram of the industrial robot current signal 4_V.

[0077] Figure 26 This is the control diagram of the industrial robot current signal 5_U.

[0078] Figure 27 This is the control diagram of the industrial robot current signal 5_V.

[0079] Figure 28 This is the control diagram of the industrial robot current signal 6_U.

[0080] Figure 29 This is the control diagram of the industrial robot current signal 6_V. DETAILED DESCRIPTION

[0081] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0082] Example 1:

[0083] This embodiment provides a method for detecting abnormal current in robot joints based on statistical process control. Figure 1 As shown, the following steps are included:

[0084] Step S1: Obtain the current signal of the industrial robot joint.

[0085] Step S2: using an autocorrelation analysis method to obtain the period of the industrial robot joint current signal;

[0086] Step S3: using a cross-correlation analysis method to calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal;

[0087] Step S4, calculating the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the period according to the period obtained in step S2;

[0088] Step S5: Perform abnormality detection using a statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal.

[0089] In this embodiment, step S1 is specifically as follows:

[0090] The U and V phase cables in the industrial robot control cabinet pass through the current transformer to collect the U and V phase currents of each joint. The industrial robot is a six-axis serial robot, and the current signals of the six joints are collected simultaneously. The current signals of the normal state and the fault state are collected with a sampling frequency of 10kHz and a sampling point number of 200,000. At this time, the 12-dimensional current signal of the industrial robot in the normal state and the 12-dimensional current signal of the industrial robot in the fault state can be obtained. The specific data of the current signal are shown in Tables 1 and 2, and the time domain diagram of the current signal is shown in Table 1. Figure 2-13 In this embodiment, the current reference signal under normal state and the current monitoring signal under fault state are selected to verify their effectiveness.

[0091] Table 1 Joint current reference signal data of industrial robots in normal state

[0092]

[0093] Table 2 Joint current monitoring signal data of industrial robot under fault condition

[0094]

[0095] In this embodiment, step S2 is specifically as follows:

[0096] Calculate the autocorrelation function of the industrial robot joint current signal, then normalize the sequence length to eliminate the influence of time lag, and finally calculate the index distance between the maximum peaks of the industrial robot joint current signal autocorrelation function to obtain the number of data points in one cycle of the industrial robot joint current signal and the time of one cycle of the industrial robot joint current signal. The calculated autocorrelation image is as follows: Figure 14 As shown. The calculated distance between the two peaks is 28627 data points (i.e., the period of the industrial robot joint current signal is 2.8627 seconds). That is, the number of period points of the industrial robot joint current signal 1_U in normal state is 28627 data points. The data of the industrial robot joint current signal 1_U in normal state is segmented by the calculated number of period points, as shown in the figure. Figure 15 As shown in the figure, analysis shows that the collected joint current signal 1_U for the industrial robot in its normal state contains six complete cycles of current signals, with the seventh cycle being incomplete. Subsequent research will be conducted based on the six complete cycles of the joint current signal 1_U for the industrial robot in its normal state. Similarly, calculations of the joint current signals for the industrial robot in normal and faulty states in other dimensions yielded the same result: the number of joint current signal cycles is 28,627. Subsequent research will also be conducted based on the six complete cycles of the joint current signal for the industrial robot in its normal state.

[0097] In this embodiment, step S3 is specifically as follows:

[0098] When analyzing and comparing the joint current monitoring signal of an industrial robot in a faulty state and the joint current reference signal of an industrial robot in a normal state, there will be a phase difference between the two signals, which makes analysis inconvenient. In order to eliminate the phase difference between the joint current monitoring signal of an industrial robot and the joint current reference signal of an industrial robot in a normal state, the joint current monitoring signal of an industrial robot and the joint current reference signal of an industrial robot in a normal state are cross-correlated to find the lag difference between the signals. Among them, the cross-correlation function The calculation expression is as follows:

[0099]

[0100] Where f(x) and g(x) represent the joint current monitoring signal and the normal joint current reference signal of the industrial robot, and m represents the phase difference between the signals. After calculating the phase difference between the two signals using a cross-correlation function, the joint current monitoring signal is phase-shifted based on the calculated phase difference, using the normal joint current reference signal as a reference. This phase difference is eliminated.

[0101] like Figure 16 As shown. By comparing the time domain diagrams, it is found that there is a phase difference between the joint current reference signal of the industrial robot in normal state and the joint current monitoring signal of the industrial robot in fault state, which makes it inconvenient to analyze. Through the cross-correlation analysis method, it is calculated that the phase difference between the joint current reference signal 1_U of the industrial robot in normal state and the joint current monitoring signal 1_U of the industrial robot in fault state is 5294 (that is, the phase time difference between the joint current signal 1_U of the industrial robot in normal state and the joint current monitoring signal 1_U of the industrial robot in fault state is 0.5294 seconds). That is, taking the joint current reference signal 1_U of the industrial robot in normal state as the benchmark, the phase of the joint current monitoring signal 1_U of the industrial robot in fault state is shifted to the left by 5294 data points, as shown Figure 17As shown in the figure, the time domain phase difference between the industrial robot's normal joint current reference signal 1_U and the faulty joint current monitoring signal 1_U is eliminated, facilitating subsequent analysis. After eliminating the phase difference, the time domain images of the normal joint current reference signal and the faulty joint current monitoring signal can be compared to detect abnormalities and perform qualitative analysis. The phase differences between the normal and faulty joint current signals "1_V," "2_U," "2_V," "3_U," "3_V," "4_U," "4_V," "5_U," "5_V," "6_U," and "6_V" are 5291, 5329, 5328, 5056, 5057, -944, -944, -10566, -10566, 1863, and 1863, respectively. Similarly, shifting is performed based on the phase point difference to eliminate the time domain phase difference, allowing for time domain image comparison and qualitative analysis.

[0102] In this embodiment, the calculation results of step S4 are shown in Table 3, where S1 to S6 represent the first to sixth cycles.

[0103] Table 3 Total energy within the period of the industrial robot's normal state reference current signal and fault state joint monitoring current signal

[0104] S1 S2 S3 S4 S5 S6 Reference signal 60922.1 57793.6 64101.6 70648.9 61572.4 58183.5 Monitoring signal 55817 67459.1 61972.2 56103.3 65147.4 64068.4

[0105] In this embodiment, step S5 specifically includes the following steps:

[0106] Step A1: Calculate the upper and lower control limits of the control chart based on the total energy of the joint current reference signal of the industrial robot in the normal state. The calculation formula is as follows:

[0107]

[0108]

[0109]

[0110]

[0111] Among them, t i represents the energy of the i-th reference signal, represents the average total energy within the period of the industrial robot's normal joint current reference signal, σ represents the standard deviation of the total energy within the period of the industrial robot's normal joint current reference signal, UCL represents the upper control limit, and LCL represents the lower control limit. Ultimately, the upper control limit (UCL) was calculated to be 75198.45, and the lower control limit (LCL) was calculated to be 49208.93.

[0112] Step A2: determine whether the total energy of the joint current monitoring signal of the industrial robot in the fault state is between the upper control limit and the lower control limit. If so, determine that there is no abnormality in the industrial robot; if not, determine that there is an abnormality in the industrial robot.

[0113] The control diagram of the joint monitoring current signal "1_V" "2_U" "2_V" "3_U" "3_V" "4_U" "4_V" "5_U" "5_V" "6_U" "6_V" in the fault state of the industrial robot is as follows Figures 18-29 As shown. Analysis shows that the joint monitoring current signals "2_V", "4_U", "4_V", "5_U", "5_V", "6_U", and "6_V" of the industrial robot in the fault state all exceed the control upper limit, and the industrial robot is abnormal. After on-site inspection, it was found that the second joint of the industrial robot had an abnormality and the monitored current signal was clipped due to the problem of sensor setting. Due to the transmission relationship of joint movement, the fluctuation caused by the fault amplified the fault signal and transmitted it to axes 4, 5, and 6. Therefore, the industrial robot joint current signal anomaly detection method based on statistical process control proposed in this embodiment successfully completed the anomaly detection of the industrial robot.

[0114] Example 2:

[0115] This embodiment provides a robot joint current anomaly detection device based on statistical process control, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to implement the following method when executing the computer program:

[0116] Step S1, obtaining the current signal of the industrial robot joint;

[0117] Step S2: using an autocorrelation analysis method based on the current signal of the industrial robot joint, obtaining the period of the industrial robot joint current signal;

[0118] Step S3: Calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal using a cross-correlation analysis method based on the current signal of the industrial robot joint;

[0119] Step S4, calculating the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the cycle based on the cycle obtained in step S2 and the current signal with the phase difference eliminated in step S3;

[0120] Step S5: Calculate a control threshold range using a statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal, and perform abnormality judgment based on the control threshold range.

[0121] Example 3:

[0122] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, it implements the robot joint current abnormality based on statistical process control as mentioned in Example 1 of the present invention. Any combination of one or more computer-readable media can be used. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device.

[0123] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for detecting abnormal current in robot joints based on statistical process control, characterized in that: The following steps are involved: S1, obtain the current signal of the industrial robot joint; S2. Obtain the period of the current signal of the industrial robot joint using the autocorrelation analysis method according to the current signal of the industrial robot joint; S3. Calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal using a cross-correlation analysis method based on the industrial robot joint current signal; S4. Calculate the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the period based on the period obtained in step S2 and the current signal with the phase difference eliminated in step S3; S5. Calculate a control threshold range using a statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal, and perform abnormality judgment based on the control threshold range; The specific steps of the statistical process control method used in step S5 are as follows: A1. Based on the total energy of the industrial robot joint current reference signal, calculate the upper and lower control limits of the control chart. The calculation formula is as follows: Among them, t i represents the energy of the i-th reference signal, represents the average value of the total energy within the period of the industrial robot's normal joint current reference signal, σ represents the standard deviation of the total energy within the period of the industrial robot's normal joint current reference signal, UCL represents the upper control limit, and LCL represents the lower control limit; A2. Determine whether the total energy of the joint current monitoring signal of the industrial robot is between the upper control limit and the lower control limit. If so, determine that there is no abnormality in the industrial robot; if not, determine that there is an abnormality in the industrial robot.

2. The method for detecting abnormal current in robot joints based on statistical process control according to claim 1, characterized in that: In step S1, the method for obtaining the current signal is to pass the U and V phase cables in the industrial robot control cabinet through the current transformer to collect the U and V phase currents of each joint.

3. The method for detecting abnormal current in robot joints based on statistical process control according to claim 1, characterized in that: The specific steps of the autocorrelation analysis method in step S2 are to obtain the autocorrelation function of the industrial robot joint current signal, then normalize the sequence length to eliminate the influence of time lag, and finally obtain the index distance between the maximum peaks of the autocorrelation function of the industrial robot joint current signal to obtain the number of data points of one cycle of the industrial robot joint current signal and the time of one cycle of the industrial robot joint current signal; The calculation expression of the autocorrelation function is as follows: Where h is the order, μ is the mean of the sequence, and x is the time series of the input signal.

4. The method for detecting abnormal current in robot joints based on statistical process control according to claim 1, characterized in that: Cross-correlation function of the cross-correlation analysis method in step S3 The calculation expression is as follows: Where f(x) represents the industrial robot joint current monitoring signal, g(x) represents the industrial robot joint current reference signal in normal state, and m represents the phase difference between the signals.

5. A robot joint current anomaly detection device based on statistical process control, characterized in that: The invention comprises a memory and a processor; the memory is used to store a computer program; the processor is used to implement the following method when executing the computer program: S1, obtain the current signal of the industrial robot joint; S2. Obtain the period of the current signal of the industrial robot joint using the autocorrelation analysis method according to the current signal of the industrial robot joint; S3. Calculate and eliminate the phase difference between the industrial robot joint current monitoring signal and the reference signal using a cross-correlation analysis method based on the industrial robot joint current signal; S4. Calculate the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal within the period based on the period obtained in step S2 and the current signal with the phase difference eliminated in step S3; S5. Calculate a control threshold range using a statistical process control method based on the total energy of the industrial robot joint current monitoring signal and the total energy of the industrial robot joint current reference signal, and perform abnormality judgment based on the control threshold range; The specific steps of the statistical process control method used in step S5 are as follows: A1. Based on the total energy of the industrial robot joint current reference signal, calculate the upper and lower control limits of the control chart. The calculation formula is as follows: Among them, t i represents the energy of the i-th reference signal, represents the average value of the total energy within the period of the industrial robot's normal joint current reference signal, σ represents the standard deviation of the total energy within the period of the industrial robot's normal joint current reference signal, UCL represents the upper control limit, and LCL represents the lower control limit; A2. Determine whether the total energy of the joint current monitoring signal of the industrial robot is between the upper control limit and the lower control limit. If so, determine that there is no abnormality in the industrial robot; if not, determine that there is an abnormality in the industrial robot.

6. The device for detecting abnormal current in robot joints based on statistical process control according to claim 5, characterized in that: In step S1, the method for obtaining the current signal is to pass the U and V phase cables in the industrial robot control cabinet through the current transformer to collect the U and V phase currents of each joint.

7. The device for detecting abnormal current in robot joints based on statistical process control according to claim 5, characterized in that: The specific steps of the autocorrelation analysis method in step S2 are to obtain the autocorrelation function of the industrial robot joint current signal, then normalize the sequence length to eliminate the influence of time lag, and finally obtain the index distance between the maximum peaks of the autocorrelation function of the industrial robot joint current signal to obtain the number of data points of one cycle of the industrial robot joint current signal and the time of one cycle of the industrial robot joint current signal; The calculation expression of the autocorrelation function is as follows: Where h is the order, μ is the mean of the sequence, and x is the time series of the input signal.

8. The device for detecting abnormal current in robot joints based on statistical process control according to claim 4, characterized in that: Cross-correlation function of the cross-correlation analysis method in step S3 The calculation expression is as follows: Where f(x) represents the industrial robot joint current monitoring signal, g(x) represents the industrial robot joint current reference signal in normal state, and m represents the phase difference between the signals.

Images