Low-voltage high-power multi-branch motor fault test method based on data analysis

Abstract

The invention relates to the technical field of multi-branch motor fault testing, in particular to a low-voltage high-power multi-branch motor fault testing method based on data analysis, and the method comprises the steps: collecting AC impedance data and current data; acquiring an alternating current impedance abnormal coefficient of each branch motor at each acquisition moment; acquiring the principal component direction of the current data in each acquisition time period; acquiring a current fluctuation coefficient of each branch motor; acquiring a data correlation coefficient between each branch motor and other branch motors; obtaining a motor abnormal coefficient of each branch motor based on the data correlation coefficient; and performing fault testing on the multi-branch motor based on the motor abnormal coefficient. According to the invention, the fault test precision of the multi-branch motor is improved.

Description

Technical Field

[0001] The present application relates to the technical field of multi-branch motor fault testing, and in particular to a low-voltage, high-power multi-branch motor fault testing method based on data analysis. Background Art

[0002] With the continuous development of industrial automation, low-voltage, high-power, multi-branch motors have been widely used in various industrial applications, such as fans, water pumps, compressors, etc. However, with the increase of usage time and changes in load conditions, these motors are prone to various faults during operation, such as winding short circuit, insulation aging, mechanical wear, etc., which seriously affect the reliability and economy of the equipment. Traditional fault detection methods mainly rely on experience and regular maintenance, and often cannot monitor the operating status of the motor in real time, resulting in delayed fault detection and increased maintenance costs. With the development of data acquisition technology and data analysis methods, the operating status and fault characteristics of the motor can be effectively identified by real-time monitoring and analysis of various data such as current, voltage, temperature, vibration, etc. during the operation of the motor.

[0003] In the prior art, an AC impedance tester is usually used to detect winding short circuit faults. The AC impedance tester applies an AC signal of known frequency and amplitude to the device under test and measures the device's response to the signal. If the motor has multiple windings, the AC impedance between two windings is measured to detect whether there is a short circuit. However, because the motor behaves differently under different load conditions, if the measurement is performed while the motor is running, it will be affected by the load change, thereby affecting the accuracy of the measurement result. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a low-voltage, high-power, multi-branch motor fault testing method based on data analysis to solve the existing problems.

[0005] The low-voltage, high-power, multi-branch motor fault testing method based on data analysis of the present application adopts the following technical solutions: An embodiment of the present application provides a low-voltage, high-power, multi-branch motor fault testing method based on data analysis, the method comprising the following steps: Collect the AC impedance data and current data of each branch motor at each collection time; Based on the average of the differences between the AC impedance data of each branch motor and other branch motors, the AC impedance abnormality coefficient of each branch motor at each acquisition time is obtained; Obtaining the principal component direction of the current data in each acquisition time period; Based on the main component direction of the current data of each branch motor and the difference between the current data and the average of all current data, the current turbulence coefficient of each branch motor is obtained; Divide the current data sequence of each branch motor; Based on the difference between the average conditions of the AC impedance abnormality coefficient, the difference between the current turbulence coefficients and the correlation between the current data, the data correlation coefficient between each branch motor and other branch motors is obtained; Obtain the motor abnormality coefficient of each branch motor based on the data correlation coefficient; Perform fault testing on multi-branch motors based on motor abnormality coefficients.

[0006] Furthermore, the calculation formula of the AC impedance anomaly coefficient is: ; In the formula, Indicates the number of The AC impedance anomaly coefficient of each branch motor; Indicates the number of AC impedance data of each motor, It represents the AC impedance data of the rth branch motor except the jth branch motor at each acquisition time. Indicates the number of branch motors, Represents the normalization function.

[0007] Furthermore, the collection time period is a time period consisting of a preset number of consecutive collection moments.

[0008] Furthermore, the method for acquiring the principal component direction is: for the current data in each acquisition time period, a principal component analysis algorithm is used to process the current data in the acquisition time period to acquire the principal component direction of the current data in each acquisition time period.

[0009] Furthermore, the calculation formula of the current turbulence coefficient is: ; In the formula, Indicates The current turbulence coefficient of each branch motor; Indicates The first branch motor The principal component direction of the current data for each acquisition period, is 180 degrees, represents an exponential function with a natural constant as the base, and m represents the total number of acquisition time periods; Indicates The i-th current data of the branch motor, Indicates The average value of all current data of the branch motors, Indicates the total number of current data.

[0010] Furthermore, the current data sequence is a sequence obtained by sorting all current data of the branch motors in positive time sequence.

[0011] Furthermore, the calculation formula of the data correlation coefficient is: ; In the formula, Indicates The branch motor and the The data correlation coefficient between the branch motors; e is a natural constant, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The current turbulence coefficient of each branch motor, Indicates The current turbulence coefficient of each branch motor, Indicates The branch motor and the The correlation coefficient between the current data series of the branch motors.

[0012] Further, the method for obtaining the motor abnormality coefficient is: Obtain each cluster based on the data correlation coefficient; Calculate the difference between the mean of all data correlation coefficients and the minimum of all data correlation coefficients, calculate the product of the difference and the number of times each branch motor appears in the cluster with the minimum mean of the data correlation coefficient, and use the normalized value of the product as the motor abnormality coefficient of each branch motor.

[0013] Furthermore, the method of obtaining each clustering cluster based on the data correlation coefficient includes: for all branch motors, using a clustering algorithm to cluster all branch motors, taking the inverse of the data correlation coefficient between any two branch motors as the metric distance between any two branch motors, and obtaining each clustering cluster.

[0014] Furthermore, the fault testing of multiple branch motors based on motor abnormality coefficients includes: when the motor abnormality coefficient of a branch motor is greater than a preset motor abnormality threshold, treating the branch motor as a faulty branch motor.

[0015] This application has at least the following beneficial effects: The present application collects the AC impedance data and current data of each branch motor at each collection moment. In order to reflect the abnormal situation of the AC impedance of each branch motor, an AC impedance abnormality coefficient is constructed based on the AC impedance data. In order to reflect the stability of the current data of each branch motor, a current turbulence coefficient is constructed based on the current data. Then, a data correlation coefficient is constructed to reflect the correlation between the data of each branch motor and other branch motors. The motor abnormality coefficient of each branch motor is obtained according to the data correlation coefficient, and fault tests are performed on multi-branch motors, thereby improving the accuracy of multi-branch motor fault tests. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flow chart of a low-voltage, high-power, multi-branch motor fault testing method based on data analysis provided in one embodiment of the present application; Figure 2 A flow chart for obtaining a motor abnormality coefficient is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of the low-voltage, high-power, multi-branch motor fault testing method based on data analysis proposed in the present application in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0020] The specific scheme of the low-voltage, high-power, multi-branch motor fault testing method based on data analysis provided by the present application is described in detail below with reference to the accompanying drawings.

[0021] An embodiment of the present application provides a low-voltage, high-power, multi-branch motor fault testing method based on data analysis. Specifically, the following low-voltage, high-power, multi-branch motor fault testing method based on data analysis is provided. Please refer to Figure 1 , the method comprises the following steps: Step S1, collecting the AC impedance data and current data of each branch motor at each collection moment.

[0022] Multi-branch permanent magnet synchronous motor has The traditional full circumference stator is divided into (n≥3) branch motor modules with completely identical motors. In this embodiment, the value of n is 5. The implementer can select other values ​​according to the actual situation. Each branch module shares a rotor, and each branch is independently powered by a general low-voltage inverter. Among them, each branch module is independent of each other and can be put into operation in different combinations of branches at will.

[0023] Furthermore, each branch motor has a winding, and the present application uses an AC impedance tester to obtain the AC impedance data of each branch motor at each collection moment; an ammeter is used to obtain the current data of each branch motor at each collection moment. In this embodiment, the interval between each collection moment is 1s, and the implementer can select other values ​​according to actual conditions.

[0024] Step S2, based on the average of the difference between the AC impedance data of each branch motor and other branch motors, obtain the AC impedance abnormality coefficient of each branch motor at each acquisition time; obtain the principal component direction of the current data in each acquisition time period; based on the principal component direction of the current data of each branch motor and the difference between the current data and the average of all current data, obtain the current turbulence coefficient of each branch motor.

[0025] Low-voltage, high-power, multi-branch motors usually use a three-phase power supply system with multiple winding branches, which can be configured into different wiring methods as needed, such as star connection or triangle connection. Low-voltage, high-power, multi-branch motors are widely used in manufacturing, mining, transportation, water treatment and other fields due to their high efficiency and good starting characteristics, driving various mechanical equipment such as pumps, fans, conveyors and compressors. Due to their high power and multi-branch characteristics, this type of motor requires special attention to electrical performance and maintenance management during operation to ensure its safe and reliable operation and avoid production stagnation and economic losses caused by failures.

[0026] Because it adopts a modular combined stator structure, it has a significant change compared to the traditional integral motor. The annular core is composed of a certain number of fan-shaped pancake-shaped module motors. According to the actual needs of process, materials and performance, the spacing between adjacent unit motors can be appropriately adjusted. Due to the use of modular winding structure and reverse embedding method, adjacent stator units can be mechanically decoupled, which greatly enhances the flexibility of motor assembly and maintenance and improves the overall reliability of the motor system.

[0027] Winding short circuit faults not only lead to the decline of motor performance, but also may cause equipment overheating, damage and even safety accidents, posing a serious threat to the stability and continuity of the production line. In addition, the occurrence of short circuit faults may also cause current surges in the power grid, affecting the normal operation of other equipment. Therefore, by regularly monitoring the electrical characteristics of the windings, such as impedance changes, temperature rise, etc., potential faults can be discovered in time and the service life of the motor can be extended.

[0028] According to the above analysis, in order to reflect the abnormal situation of the AC impedance of each branch motor, the AC impedance abnormality coefficient of each branch motor is calculated based on the average of the difference between the AC impedance data of each branch motor and other branch motors. The calculation formula is as follows: ; In the formula, Indicates the number of The AC impedance anomaly coefficient of each branch motor; Indicates the number of AC impedance data of each motor, It represents the AC impedance data of the rth branch motor except the jth branch motor at each acquisition time. Indicates the number of branch motors, Represents the normalization function.

[0029] It should be noted that when a branch motor has a short circuit problem, the AC impedance of the branch motor will decrease. The AC impedance of each branch motor is significantly different from that of other branch motors. The larger the value of , the larger the AC impedance abnormality coefficient obtained by the branch motor at each acquisition time, and the greater the possibility of a short circuit abnormality; conversely, the smaller the AC impedance abnormality coefficient obtained by the branch motor at each acquisition time, the smaller the possibility of a short circuit abnormality.

[0030] Furthermore, during the long-term operation of the motor winding, the load change of the motor will affect its working state. When the load is too large, the motor core may enter the magnetic saturation state. In the magnetic saturation state, the inductive reactance of the motor will decrease, resulting in a change in the AC impedance. When the load increases, the current required by the motor will also increase. The increase in current will directly affect the resistance and reactance of the motor winding, resulting in a change in the total AC impedance. For inductive loads, the increase in current will also affect the reaction time and phase angle of the motor, thereby changing the impedance characteristics.

[0031] Therefore, when identifying a short-circuit fault in a motor winding, the change in the motor load will cause the AC impedance to change. It is inaccurate to judge whether the motor has a short-circuit fault based only on the AC impedance. Therefore, it is also necessary to analyze the change in the motor current and identify the abnormal situation based on the change in the current data.

[0032] A time period consisting of a preset number of continuous acquisition moments is used as each acquisition time period. In this embodiment, the preset number is 10, and the implementer can select other values ​​according to actual conditions. For the current data of each acquisition time period, the principal component analysis (PCA) algorithm is used to process the current data of the acquisition time period to obtain the principal component direction of the current data of each acquisition time period. The principal component analysis (PCA) algorithm is a well-known technology and will not be described in detail in this embodiment.

[0033] Furthermore, in order to reflect the stability of the current data of each branch motor, the current turbulence coefficient of each branch motor is obtained based on the main component direction of the current data of each branch motor and the difference between the current data and the average of all current data. The calculation formula is: ; In the formula, Indicates The current turbulence coefficient of each branch motor; Indicates The first branch motor The principal component direction of the current data for each acquisition period, is 180 degrees, represents an exponential function with a natural constant as the base, and m represents the total number of acquisition time periods; Indicates The i-th current data of the branch motor, Indicates The average value of all current data of the branch motors, Indicates the total number of current data.

[0034] It should be noted that Indicates the change in the direction of the principal component in each acquisition time period. When the principal component direction is positive, the smaller the difference, the current data is in a positive rising state, so the motor load is greater; if the principal component direction is negative, the larger the difference, the data is in a falling state, so the motor load is smaller; Indicates the value of angle converted to radians; Indicates the change trend of each segment in the entire current data sequence. The larger the value, the greater the change degree of the current data, and therefore the more unstable it is; Indicates The current data of the branch motor The difference between the amplitude of a data point and the mean value of the current data, It indicates the degree of change of the current data. The greater the degree of change, the more unstable the current data is, and the larger the current turbulence coefficient obtained at this time; conversely, the smaller the current turbulence coefficient obtained is.

[0035] Based on the stability of the current data changes of different branch motors obtained from the above calculation, we then compare the correlation between the current data changes of different motors. Because under normal circumstances, different branch motors work at the same time, and their states are stable and similar. When one of the motors has a problem, its current and AC impedance will change; but when the load of the motor increases, the current of each motor will change, and the change is relatively uniform.

[0036] Step S3, divide the current data sequence of each branch motor; obtain the data correlation coefficient between each branch motor and other branch motors based on the difference between the average conditions of the AC impedance abnormality coefficient, the difference between the current turbulence coefficients and the correlation between the current data; obtain the motor abnormality coefficient of each branch motor based on the data correlation coefficient.

[0037] For each branch motor, all current data of the branch motor are sorted in chronological order to obtain a current data sequence of each branch motor.

[0038] Further, according to the above analysis, in order to reflect the correlation between the data of each branch motor and other branch motors, based on the difference between the average conditions of the AC impedance abnormality coefficient, the difference between the current turbulence coefficients and the correlation between the current data, the data correlation coefficient between any two branch motors is calculated. The calculation formula is: ; In the formula, Indicates The branch motor and the The data correlation coefficient between the branch motors; e is a natural constant, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The current turbulence coefficient of each branch motor, Indicates The current turbulence coefficient of each branch motor, Indicates The branch motor and the The correlation coefficient between the current data sequences of the branch motors; wherein, this embodiment selects the Pearson correlation coefficient as the measurement method of the correlation coefficient, and the implementer can select other correlation coefficients according to actual conditions.

[0039] It should be noted that Represents the difference in AC impedance of the two branch motors. If there is a large difference between the AC impedance of the two branch motors, then the working conditions of the motors will be very different. Therefore, the correlation is small, so it is used here. To represent the difference in state and amplify it exponentially; Indicates the difference between the current stability of the two branch motors. The greater the difference, the greater the difference in the operating states between the different motors. At this time, the correlation between the two branch motors is smaller. Indicates The branch motor and the The Pearson correlation coefficient of the branch motor current data describes the correlation between the current data sequences of the two branch motors. The smaller the correlation, the less similar the changes in the current data sequences of the two branch motors, and the smaller the value of the data correlation coefficient obtained at this time; otherwise, the value of the data correlation coefficient obtained is larger.

[0040] Furthermore, when the data correlation coefficient between any two branch motors is small, there is a large difference between the state of the branch motor and other motors. The reason for the difference is that there is a large difference between the current data and the AC impedance. When the difference is relatively large, it means that the current motor has a short circuit fault. Therefore, the abnormal branch is located based on the correlation between the motor states obtained from the above calculation.

[0041] Because when a motor fails, its correlation with other motors will be small, therefore, according to the above analysis, for all branch motors, a clustering algorithm is used to cluster all branch motors, and the inverse of the data correlation coefficient between any two branch motors is used as the metric distance between any two branch motors to obtain each cluster cluster. Among them, the clustering algorithm selected in this embodiment is the DBSCAN clustering algorithm, and the implementer can select other clustering algorithms according to actual conditions.

[0042] Furthermore, according to the cluster division results, the motor abnormality coefficient of each branch motor is calculated, and the calculation formula is: ; In the formula, Indicates The motor abnormality coefficient of each branch motor, represents the mean of all data correlation coefficients, represents the minimum value of all data correlation coefficients, Indicates the cluster with the smallest mean value of the data correlation coefficient. The number of times a motor appears, is a normalized function. Among them, the flow chart of obtaining the motor abnormality coefficient is as follows: Figure 2 shown.

[0043] It should be noted that, for each branch motor, when the branch motor appears more times in the cluster with the smallest mean value of the data correlation coefficient, it means that the correlation between the motor and other motors is smaller, the branch motor is more abnormal, and the possibility of failure is greater; when the difference between the mean value of all data correlation coefficients and the minimum value of all data correlation coefficients is large, the possibility of a faulty branch motor is greater, and the motor abnormality coefficient of each motor is larger; conversely, the motor abnormality coefficient is smaller.

[0044] Step S4, performing a fault test on the multi-branch motor based on the motor abnormality coefficient.

[0045] According to the abnormality of the motor obtained by the above calculation, a preset motor abnormality threshold is set. In this embodiment, the value of the preset motor abnormality threshold is 0.7, and the implementer can select other values ​​according to the actual situation; specifically, when the motor abnormality coefficient of the branch motor is greater than the preset motor abnormality threshold, it means that there is a large difference in current and AC impedance between the current motor and other motors, and it is because of the fault caused by winding short circuit, and the branch motor is regarded as a faulty branch motor.

[0046] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of the present application are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0047] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0048] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A low-voltage, high-power, multi-branch motor fault testing method based on data analysis, characterized in that: The method comprises the following steps: Collect the AC impedance data and current data of each branch motor at each collection time; Based on the average of the differences between the AC impedance data of each branch motor and other branch motors, the AC impedance abnormality coefficient of each branch motor at each acquisition time is obtained; Obtaining the principal component direction of the current data in each acquisition time period; Based on the main component direction of the current data of each branch motor and the difference between the current data and the average of all current data, the current turbulence coefficient of each branch motor is obtained; Divide the current data sequence of each branch motor; Based on the difference between the average conditions of the AC impedance abnormality coefficient, the difference between the current turbulence coefficients and the correlation between the current data, the data correlation coefficient between each branch motor and other branch motors is obtained; Obtain the motor abnormality coefficient of each branch motor based on the data correlation coefficient; Perform fault testing on multi-branch motors based on motor abnormality coefficients.

2. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The calculation formula of the AC impedance anomaly coefficient is: ; In the formula, Indicates the number of The AC impedance anomaly coefficient of each branch motor; Indicates the number of AC impedance data of each motor, It represents the AC impedance data of the rth branch motor except the jth branch motor at each acquisition time. Indicates the number of branch motors, Represents the normalization function.

3. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The collection time period is a time period consisting of a preset number of consecutive collection moments.

4. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The method for acquiring the principal component direction is: for the current data in each acquisition time period, a principal component analysis algorithm is used to process the current data in the acquisition time period to acquire the principal component direction of the current data in each acquisition time period.

5. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The calculation formula of the current turbulence coefficient is: ; In the formula, Indicates The current turbulence coefficient of each branch motor; Indicates The first branch motor The principal component direction of the current data for each acquisition period, is 180 degrees, represents an exponential function with a natural constant as the base, and m represents the total number of acquisition time periods; Indicates The i-th current data of the branch motor, Indicates The average value of all current data of the branch motors, Indicates the total number of current data.

6. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The current data sequence is a sequence obtained by sorting all current data of the branch motors in positive time sequence.

7. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The calculation formula of the data correlation coefficient is: ; In the formula, Indicates The branch motor and the The data correlation coefficient between the branch motors; e is a natural constant, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The average value of all AC impedance anomaly coefficients of the branch motors, Indicates The current turbulence coefficient of each branch motor, Indicates The current turbulence coefficient of each branch motor, Indicates The branch motor and the The correlation coefficient between the current data series of the branch motors.

8. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The method for obtaining the motor abnormality coefficient is: Obtain each cluster based on the data correlation coefficient; Calculate the difference between the mean of all data correlation coefficients and the minimum of all data correlation coefficients, calculate the product of the difference and the number of times each branch motor appears in the cluster with the minimum mean of the data correlation coefficient, and use the normalized value of the product as the motor abnormality coefficient of each branch motor.

9. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 8, characterized in that: The method of obtaining each clustering cluster based on the data correlation coefficient includes: for all branch motors, using a clustering algorithm to cluster all branch motors, taking the inverse of the data correlation coefficient between any two branch motors as the metric distance between any two branch motors, and obtaining each clustering cluster.

10. The low-voltage, high-power, multi-branch motor fault testing method based on data analysis as claimed in claim 1, characterized in that: The method of performing fault testing on multiple branch motors based on motor abnormality coefficients includes: when the motor abnormality coefficient of a branch motor is greater than a preset motor abnormality threshold, treating the branch motor as a faulty branch motor.

Images