A motor operation fault analysis method based on electrical variable analysis

Abstract

The present invention discloses a motor operation fault analysis method based on electric variable analysis, which specifically relates to the technical field of motor fault analysis, including: collecting motor operation data, dividing the collected motor operation data into a plurality of areas according to operation conditions, building a motor virtual simulation model corresponding to each operation condition, obtaining abnormal information by comparing the collected operation data with simulation prediction results, comparing the simulation prediction results with actual dependent variables, and summarizing the abnormal information of each operation condition; inputting the abnormal information of each operation condition into a corresponding fault diagnosis model, outputting a fault type probability matrix and a preliminary fault type set, mutually verifying the output results of each operation condition, and judging whether the preliminary fault type set needs to be corrected, thereby effectively solving the problem that the existing motor operation fault analysis is not fast and accurate enough, and improving the accuracy and reliability of fault diagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of motor fault analysis, and more specifically, to a motor operation fault analysis method based on electrical variable analysis. Background Art

[0002] Motor operation fault analysis is of great significance in industrial production. The occurrence of motor failure will cause the production line to stop for too long, affecting product quality and production efficiency. Through fault analysis, problems can be quickly located and solved, downtime can be reduced, production efficiency can be improved, and then production efficiency can be improved and maintenance costs can be reduced.

[0003] In the Chinese invention patent with application publication number CN118503652A, a fault monitoring method and system for generator carbon brushes is disclosed, which includes the following steps: establishing a generator carbon brush operation data monitoring system; based on the generator carbon brush operation data monitoring system, obtaining the historical carbon brush operation data of the faulty generator for pre-screening and processing, and obtaining the training carbon brush fault characteristic variables; establishing an initial carbon brush fault analysis model, inputting the training carbon brush fault characteristic variables into the initial carbon brush fault analysis model for accuracy training, and obtaining the optimal carbon brush fault analysis model after training; continuing to obtain the current carbon brush operation data in real time based on the generator carbon brush operation data monitoring system, inputting the original characteristic variables corresponding to the current carbon brush operation data into the optimal carbon brush fault analysis model, and analyzing and judging and outputting the fault monitoring analysis results and replacement analysis results of the current carbon brush operation data based on the optimal carbon brush fault analysis model. The problem that it is difficult to timely discover and analyze the early warning carbon brush fault in the prior art is solved.

[0004] Motor operation fault analysis is a key link to ensure the stable operation of the motor. Common methods in the prior art include observation method, temperature detection method, vibration and noise detection method, speed detection method and insulation resistance detection method. The observation method mainly detects obvious damage and abnormal phenomena by checking the surface of the motor and the surrounding environment. The temperature detection method uses tools such as infrared thermometers to monitor the temperature of the motor to determine whether there is an overheating problem. The vibration and noise detection method uses a stethoscope or sound level meter to detect abnormal vibration and noise during motor operation to find potential mechanical failures. The speed detection method measures the input power and output torque of the motor to calculate the speed and determine whether there is an internal mechanical failure. The insulation resistance detection method uses an insulation resistance meter to measure the insulation resistance of the motor to find insulation problems.

[0005] However, in actual use, it still has many shortcomings, such as the problem that the existing motor operation fault analysis is not fast and accurate enough. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a motor operation fault analysis method based on electrical variable analysis to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a motor operation fault analysis method based on electrical variable analysis, comprising the following steps:

[0008] Step S11, collecting motor operation data, collecting motor operation data at a set sampling frequency, the motor operation data including: motor operation independent variables and motor operation dependent variables;

[0009] Step S12, operation data division: used to divide the collected motor operation data into several areas according to the operation conditions, assuming that there are n operation conditions, and i represents the sequence number of the operation conditions;

[0010] Step S13, building a motor virtual simulation model corresponding to each operating condition, inputting the motor operation independent variable data of each operating condition into the corresponding motor virtual simulation model, obtaining a simulation prediction result, and obtaining abnormal information by comparing the collected operating data with the simulation prediction result, and comparing the simulation prediction result with the actual dependent variable, and summarizing the abnormal information of each operating condition;

[0011] Step S14: Input each operating condition abnormal information into the corresponding fault diagnosis model, output the fault type probability matrix and the preliminary fault type set, and record the preliminary fault type set as set A1, A2, ..., A n ;

[0012] Step S15: The output results of each operating condition are mutually verified to determine whether the preliminary fault type set needs to be corrected.

[0013] Preferably, the operating condition is obtained based on the load condition and the operating state, including:

[0014] Classified by load conditions, motor operating conditions include no-load conditions, rated load conditions, and overload conditions;

[0015] Classified by operating status, motor operating conditions include starting conditions, braking conditions and stable conditions;

[0016] Based on the permutations and combinations of load conditions and operating states, 9 operating conditions are obtained.

[0017] Preferably, the motor virtual simulation model is obtained in the following manner:

[0018] Step S21, establishing a geometric model of the motor, ensuring that the accuracy of the geometric model matches the actual size and shape of the motor;

[0019] Step S22, defining material properties: defining material properties and temperature dependence in the motor;

[0020] Step S23, setting boundary conditions and excitation sources: setting boundary conditions and excitation sources according to the actual situation of the motor; setting initial conditions for simulation;

[0021] Step S24, performing simulation calculation: after all parameters and conditions are set, performing simulation calculation to calculate the performance parameters and working state of the motor; the simulation mode includes steady-state simulation and transient simulation, the steady-state simulation is used to calculate the performance parameters of the motor under a stable working state, and the transient simulation is used to analyze the performance parameters of the motor in a transient process;

[0022] Step S25: After the simulation calculation is completed, the simulation results are analyzed and evaluated, and the simulation confidence of each operating condition is recorded, and the simulation confidence of the i-th operating condition is recorded as CL_i.

[0023] Preferably, the method for obtaining the fault type probability matrix and the preliminary fault type set comprises the following steps:

[0024] Step S31, data collection and sorting: sorting the motor operation data of each operating condition, including current, voltage, electric power, power factor, efficiency, iron loss information, torque output, speed, temperature, vibration information and air gap information between the rotor and the stator;

[0025] Step S32, fault feature library construction: establish a fault feature library containing feature parameters of various fault types under different working conditions, the fault feature library contains a detailed description of the fault type, feature parameters and corresponding threshold range information;

[0026] Step S33, characteristic parameter analysis: inputting the motor operation data of each operating condition into a fault diagnosis model, the fault diagnosis model includes:

[0027] Feature parameter extraction: extract key feature parameters from the motor operation data of each operating condition;

[0028] Feature parameter comparison: compare the extracted feature parameters with the data in the fault feature library, and use the similarity to represent the predicted probability of the fault type;

[0029] Probability judgment: Compare the predicted probability of the fault type with the corresponding preset value. If it exceeds the preset value, it is judged that the corresponding fault type exists;

[0030] Step S34, preliminary fault diagnosis: obtain a preliminary fault type set based on the results of feature parameter comparison and probability judgment, and output a fault type probability matrix and a preliminary fault type set.

[0031] Preferably, the motor operation fault analysis method includes a step of acquiring a set of motor fault types, including:

[0032] Step S41, obtaining a fault type probability matrix, and calculating a predicted probability difference or an average predicted probability difference of each fault type under different operating conditions;

[0033] Step S42: If the predicted probability difference or the average predicted probability difference of the fault type under different operating conditions is greater than the threshold, it is marked as a fault type to be verified and the next step is entered; if the probability difference of the fault type under different operating conditions is not greater than the threshold, the fault type set is not modified, and the set A1, A2, ..., A n Take the union and get the motor fault type set QA;

[0034] Step S43, sequentially correcting the predicted probabilities of the fault types to be verified in the fault type probability matrix to obtain a corrected fault type probability matrix;

[0035] Step S44: Compare the corrected fault type probability matrix with the corresponding preset value to obtain a corrected fault type set, recorded as set XA1, XA2, ..., XA n , and take the union to obtain the corrected motor fault type set QB.

[0036] Preferably, when the fault type occurs in only two operating conditions, the predicted probability difference is directly calculated; when the fault type occurs in more than two operating conditions, the maximum predicted probability is used as the first benchmark, and the differences between the remaining predicted probabilities and the first benchmark are calculated respectively to obtain the average probability difference.

[0037] Preferably, the correction method of the predicted probability of the fault type to be verified is:

[0038] Assume that there are m fault types in the motor, and j is used to represent the order of the fault types; the predicted probability of the jth fault type in the i-th operating condition is recorded as P ij ;

[0039] The predicted probability P is calculated by the following formula ij Modified probability of XP ij ,

[0040]

[0041] Among them, xs_ij represents the trend conformity of the characteristic parameters corresponding to the jth fault type in the i-th operating condition; max_xs represents the maximum value of the trend conformity of the fault type;

[0042] The trend conformity is obtained by taking the characteristic parameter with the largest prediction probability as the second benchmark and establishing a trend analysis curve based on the second benchmark; analyzing the conformity between the characteristic parameter corresponding to the jth fault type in the i-th operating condition and the trend analysis curve to obtain xs_ij.

[0043] Preferably, the motor operation fault analysis method comprises a fault diagnosis model management step,

[0044] Calculate the intersection-and-union ratio IOU_AB of the motor fault type set QA and the corrected motor fault type set QB;

[0045] By formula The reliability index GK of the fault diagnosis model is calculated;

[0046] When the reliability index of the fault diagnosis model is lower than the threshold, an early warning is issued to indicate that the current fault diagnosis model is abnormal and measures need to be taken.

[0047] Preferably, after step S13 and before step S14, the following steps are further included:

[0048] Analyze the abnormal information of each operating condition to obtain the abnormal operation coefficient and the growth rate of the abnormal operation coefficient;

[0049] When the operation abnormality coefficient and the operation abnormality coefficient growth rate exceed the corresponding threshold, the sampling frequency of the next time period is increased.

[0050] Preferably, the operation abnormality coefficient is obtained in the following manner:

[0051] According to the time occupied by the working condition in the overall operation of the motor, the matching weight coefficient of each working condition is determined to obtain the operating time T of the i-th operating condition i , through the formula Calculate and obtain the weight coefficient of the i-th operating condition;

[0052] Get the simulation parameters and actual operating parameters of each operating condition. Assume that there are R operating parameters in each operating condition, and use k to represent the sequential number of the operating parameters. The abnormal operating parameters of the i-th condition are calculated by the following formula:

[0053]

[0054] in, represents the actual parameter of the kth operating parameter of the ith operating condition, represents the simulation prediction parameter of the kth operating parameter of the ith operating condition, represents the rated parameter of the kth operating parameter in the ith operating condition;

[0055] Based on the abnormal operating parameters of each operating condition, the joint analysis is performed through the formula The operation abnormality coefficient Ya is calculated.

[0056] Technical effects and advantages of the present invention:

[0057] (1) The motor operation fault analysis method provided by the present invention realizes the refined management of the motor operation status by subdividing the motor operation data into different areas according to the working conditions; based on these area divisions, a simulation model is built to more accurately simulate the performance of the motor under different working conditions, thereby providing a reliable basis for formulating effective maintenance and management strategies; in addition, in-depth data analysis is carried out for each working condition area, which makes abnormal data easier to identify, thereby improving the accuracy of fault diagnosis; fault type analysis is carried out based on the abnormal information of each operating condition, the fault type is determined, and the output results under different working conditions are mutually verified; through the comprehensive application of working condition division, simulation model application and data analysis, the present method realizes rapid and accurate diagnosis of the motor operation status, effectively solving the problem that the existing motor operation fault analysis is not fast and accurate enough.

[0058] (2) The motor operation fault analysis method provided by the present invention intelligently adjusts the monitoring resources of each operating condition based on the operating abnormality coefficient of each operating condition, including increasing the sampling frequency of the operating condition when an abnormality is found, thereby improving the efficiency of fault analysis and effectively solving the problem of low efficiency of existing motor fault analysis and inability to intelligently adjust the sampling frequency. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of the motor operation fault analysis of the present invention;

[0060] Figure 2 This is a flow chart of a method for obtaining a corrected fault type set according to the present invention. DETAILED DESCRIPTION

[0061] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0062] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0063] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0064] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0065] Background technology: The same fault type may not manifest itself exactly the same in different operating conditions, but there are connections. By analyzing the connections and characteristic parameters, the performance characteristics of the same fault type under different conditions, as well as the similarities and differences between them, are recorded. The preliminary fault diagnosis results are mutually verified with the output results of different operating conditions. If there are inconsistencies or contradictions, corrections and confirmations are made, which helps to improve the efficiency and accuracy of motor fault analysis.

[0066] Example 1, see Figure 1 The present invention provides a flow chart of motor operation fault analysis. Figure 1 A motor operation fault analysis method based on electric variable analysis is shown, comprising the following steps:

[0067] Step S11, collecting motor operation data, collecting motor operation data at a set sampling frequency, the motor operation data including: motor operation independent variables and motor operation dependent variables;

[0068] Step S12, operation data division: used to divide the collected motor operation data into several areas according to the operation conditions, assuming that there are n operation conditions, and i represents the sequence number of the operation conditions;

[0069] Explanation, based on load conditions and operating status to obtain operating conditions, including:

[0070] Classified by load conditions, the motor operating conditions include no-load conditions, rated load conditions and overload conditions; the no-load condition refers to the motor operating without load, the rated load condition refers to the motor operating under rated load, and the overload condition refers to the motor operating in a state exceeding the rated load;

[0071] Classified by operating status, motor operating conditions include starting conditions, braking conditions and stable conditions;

[0072] Based on the load conditions and operating status, 9 operating conditions are obtained; if the rated load condition and overload condition are subdivided into more gradients, more operating conditions will be obtained;

[0073] Step S13, building a motor virtual simulation model corresponding to each operating condition, inputting the motor operation independent variable data of each operating condition into the corresponding motor virtual simulation model, obtaining a simulation prediction result, and obtaining abnormal information by comparing the collected operating data with the simulation prediction result, and comparing the simulation prediction result with the actual dependent variable, and summarizing the abnormal information of each operating condition;

[0074] Step S14: Input each operating condition abnormal information into the corresponding fault diagnosis model, output the fault type probability matrix and the preliminary fault type set, and record the preliminary fault type set as set A1, A2, ..., A n , each set represents a preliminary fault diagnosis result under an operating condition; the fault type probability matrix contains the predicted probability of each fault type under each operating condition; the fault type probability matrix is ​​a two-dimensional table, in which rows represent different fault types and columns represent different operating condition types;

[0075] Step S15: The output results of each operating condition are mutually verified to determine whether the preliminary fault type set needs to be corrected.

[0076] What needs to be further explained in the embodiments of the present invention is that the motor operation independent variables and the motor operation dependent variables are set according to actual conditions, and the embodiments of the present invention do not specifically limit this; in a possible embodiment, the motor operation independent variables may include current, voltage, electric power, power factor, efficiency, magnetic field intensity change rate, hysteresis loop area, core volume, core material resistivity, core thickness and alternating magnetic field frequency; the motor operation dependent variables may include the torque output and speed of the motor, and operating status information, and the operating status information includes temperature and vibration information.

[0077] It needs to be further explained in the embodiment of the present invention that the motor virtual simulation model is obtained in the following manner:

[0078] Step S21, establishing a geometric model of the motor: This is accomplished by importing a CAD file or manually creating a geometric model of the motor in the software; ensuring that the accuracy of the geometric model matches the actual size and shape of the motor;

[0079] Step S22, defining material properties: defining material properties in the motor, such as electrical conductivity, magnetic permeability, density, and temperature dependence, that is, considering the change of material properties with temperature, and using a deep learning model to predict material properties that change with the environment;

[0080] Step S23, setting boundary conditions and excitation sources: according to the actual situation of the motor, setting boundary conditions (such as external magnetic field, temperature distribution, mechanical constraints, ensuring that the boundary conditions match the actual working environment of the motor) and excitation sources (such as setting the motor winding current, voltage, external magnetic field); setting the initial conditions of the simulation, such as initial magnetic field distribution, initial temperature;

[0081] Step S24, performing simulation calculation: after setting all parameters and conditions, performing simulation calculation to calculate the performance parameters and working state of the motor, such as electromagnetic field distribution, current distribution, and output torque; the simulation mode includes steady-state simulation and transient simulation, the steady-state simulation is used to calculate the performance parameters of the motor under a stable working state, and the transient simulation is used to analyze the performance parameters of the motor in a transient process;

[0082] Step S25: After the simulation calculation is completed, the simulation results are analyzed and evaluated, and the simulation confidence of each operating condition is recorded, and the simulation confidence of the i-th operating condition is recorded as CL_i.

[0083] It needs to be further explained in the embodiment of the present invention that the method for obtaining the fault type probability matrix and the preliminary fault type set includes the following steps:

[0084] Step S31, data collection and sorting: sorting the motor operation data of each operating condition, including current, voltage, electric power, power factor, efficiency, iron loss information, torque output, speed, temperature, vibration information and air gap information between the rotor and the stator;

[0085] Step S32, fault feature library construction: establish a fault feature library containing feature parameters of various fault types under different working conditions, the fault feature library contains a detailed description of the fault type, feature parameters and corresponding threshold range information;

[0086] Step S33, characteristic parameter analysis: inputting the motor operation data of each operating condition into a fault diagnosis model, the fault diagnosis model includes:

[0087] Feature parameter extraction: Extract key feature parameters from the motor operation data of each operating condition, such as current fluctuation, voltage stability, power factor change, temperature anomaly, and vibration frequency;

[0088] Feature parameter comparison: compare the extracted feature parameters with the data in the fault feature library, and use the similarity to represent the predicted probability of the fault type;

[0089] Probability judgment: Compare the predicted probability of the fault type with the corresponding preset value. If it exceeds the preset value, it is judged that the corresponding fault type exists;

[0090] Step S34, preliminary fault diagnosis: obtain a preliminary fault type set based on the results of feature parameter comparison and probability judgment, and output a fault type probability matrix and a preliminary fault type set.

[0091] See also Figure 2 The flowchart of the method for obtaining the corrected fault type set is as follows:

[0092] Step S41, obtaining a fault type probability matrix, calculating the predicted probability difference or the average predicted probability difference of each fault type under different operating conditions, that is, comparing the predicted probability difference of the same fault type under different operating conditions;

[0093] Step S42: If the predicted probability difference or the average predicted probability difference of the fault type under different operating conditions is greater than the threshold, it indicates that the probability prediction of the corresponding fault type is inaccurate, and it is marked as a fault type to be verified, and the next step is entered; if the probability difference of the fault type under different operating conditions is not greater than the threshold, it indicates that the probability prediction of the fault type is accurate, and the fault type set is not modified, and the set A1, A2, ..., A n Take the union and get the motor fault type set QA;

[0094] Step S43, sequentially correcting the predicted probabilities of the fault types to be verified in the fault type probability matrix to obtain a corrected fault type probability matrix;

[0095] Step S44: Compare the corrected fault type probability matrix with the corresponding preset value to obtain a corrected fault type set, recorded as set XA1, XA2, ..., XA n , and take the union to obtain the corrected motor fault type set QB.

[0096] Explanation: When the fault type occurs in only two operating conditions, the predicted probability difference is directly calculated; when the fault type occurs in more than two operating conditions, the maximum predicted probability is used as the first benchmark, and the difference between the remaining predicted probability and the first benchmark is calculated to obtain the average probability difference.

[0097] It needs to be further explained in the embodiment of the present invention that the correction method of the predicted probability of the fault type to be verified is:

[0098] Assume that there are m fault types in the motor, and j is used to represent the order of the fault types; the predicted probability of the jth fault type in the i-th operating condition is recorded as P ij ;

[0099] The predicted probability P is calculated by the following formula ij Modified probability of XP ij ,

[0100]

[0101] Among them, xs_ij represents the trend conformity of the characteristic parameters corresponding to the jth fault type in the i-th operating condition; max_xs represents the maximum value of the trend conformity of the fault type (from 1 to m);

[0102] The trend conformity degree is obtained by taking the characteristic parameter with the largest prediction probability as the second benchmark and establishing a trend analysis curve based on the second benchmark;

[0103] Analyze the degree of conformity between the characteristic parameters corresponding to the jth fault type in the i-th operating condition and the trend analysis curve to obtain xs_ij.

[0104] Explanation: It means that among multiple characteristic parameters, the characteristic parameter with the highest prediction probability when predicting the fault type is selected as the benchmark for analysis. This benchmark represents the most significant feature of the fault type. The trend analysis curve represents the expected trend of the characteristic parameter changing with the working condition (or time). The shape, slope and intercept parameters of the trend analysis curve are determined by historical data, prediction model or expert knowledge. The degree of conformity is obtained by comparing the similarity between the actual change trend of the characteristic parameter and the expected trend. The similarity can be calculated by a variety of methods, such as calculating the correlation coefficient between the two, distance measurement (such as Euclidean distance, Manhattan distance, etc.) or goodness of fit test.

[0105] Explanation: The trend information of the same fault type under different working conditions is calculated. The trend information at least includes the change direction and change speed of the characteristic parameters, which is obtained by performing time series analysis or trend analysis on the fault characteristic parameters; the correction probability of the corresponding fault type is calculated based on the trend compliance degree.

[0106] Embodiment 2, the embodiment of the present invention provides a motor operation fault analysis method, which is different from embodiment 1 in that the motor operation fault analysis method includes a fault diagnosis model management step, specifically:

[0107] Calculate the intersection-and-union ratio IOU_AB of the motor fault type set QA and the corrected motor fault type set QB;

[0108] By formula

[0109]

[0110] The reliability index GK of the fault diagnosis model is calculated;

[0111] When the reliability index of the fault diagnosis model is lower than the threshold, an early warning is issued to indicate that the current fault diagnosis model is abnormal and measures need to be taken, such as retraining, updating the training set, and refining the classification of working conditions.

[0112] Summary: By introducing the fault diagnosis model management steps, real-time evaluation and monitoring of the fault diagnosis model performance is achieved; when the fault diagnosis model performance does not meet the standards, early warnings can be issued in time and corresponding measures can be taken, thereby improving the accuracy and reliability of motor operation fault analysis, which is of great significance for ensuring the stable operation of the motor system.

[0113] Embodiment 3, the embodiment of the present invention provides a motor operation fault analysis method, which is different from Embodiment 1 and Embodiment 2 in that after step S13 and before step S14, it also includes:

[0114] Analyze the abnormal information of each operating condition to obtain the abnormal operation coefficient and the growth rate of the abnormal operation coefficient;

[0115] When the operation abnormality coefficient and the operation abnormality coefficient growth rate exceed the corresponding threshold, the sampling frequency of the next time period is increased.

[0116] It needs to be further explained in the embodiment of the present invention that the operation abnormality coefficient is obtained in the following manner:

[0117] According to the time occupied by the working condition in the overall operation of the motor, the matching weight coefficient of each working condition is determined to obtain the operating time T of the i-th operating condition i , through the formula Calculate and obtain the weight coefficient of the i-th operating condition;

[0118] The simulation parameters and actual operating parameters of each operating condition are obtained. Assume that there are R operating parameters in each operating condition, and k represents the sequential number of the operating parameters. The abnormal operating parameters of the i-th condition are calculated by the following formula:

[0119]

[0120] in, represents the actual parameter of the kth operating parameter of the ith operating condition, represents the simulation prediction parameter of the kth operating parameter of the ith operating condition, It represents the rated parameters of the kth operating parameters in the ith working condition; the rated parameters refer to the standard parameters determined according to the performance standards and specifications when the motor is designed and manufactured;

[0121] Based on the abnormal operating parameters of each operating condition, the joint analysis is performed through the formula The operation abnormality coefficient Ya is calculated.

[0122] Summary: The current operating status of the motor is judged based on the operating abnormality coefficient and its growth rate output by the electrical variable analysis module; if the operating abnormality coefficient or growth rate exceeds the preset threshold, it means that the motor may have a fault or an unstable operating state. At this time, the sampling frequency needs to be adjusted to obtain more accurate data. By adjusting the sampling frequency, the changes in the motor's operating state can be better captured, providing more accurate data support for subsequent analysis.

[0123] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0124] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0125] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A motor operation fault analysis method based on electrical variable analysis, characterized in that: The following steps are involved: Motor operation data collection, at a set sampling frequency, collecting and obtaining motor operation data, the motor operation data including: motor operation independent variables and motor operation dependent variables; Operation data division: used to divide the collected motor operation data into several areas according to the operation conditions. Assume that there are n operation conditions, and i represents the sequence number of the operation conditions. Build a motor virtual simulation model corresponding to each operating condition, input the motor operation independent variable data of each operating condition into the corresponding motor virtual simulation model, obtain simulation prediction results, compare the collected operating data with the simulation prediction results, compare the simulation prediction results with the actual dependent variables to obtain abnormal information, and summarize the abnormal information of each operating condition; The abnormal information of each operating condition is input into the corresponding fault diagnosis model, and the fault type probability matrix and the preliminary fault type set are output. The preliminary fault type set is recorded as set A1, A2, ..., A n ; The fault diagnosis model is connected to the fault diagnosis model management step, and the fault diagnosis model management step includes: Calculate the intersection-and-union ratio IOU_AB of the motor fault type set QA and the corrected motor fault type set QB; The fault diagnosis model reliability index GK is calculated by the following formula: ； Among them, CL_i represents the simulation confidence of the i-th operating condition; When the reliability index of the fault diagnosis model is lower than the threshold, an early warning is issued to indicate that the current fault diagnosis model is abnormal and measures need to be taken; The output results of each operating condition are mutually verified to determine whether the preliminary fault type set needs to be corrected; The mutual verification method is: obtaining a fault type probability matrix, and calculating the predicted probability difference or average value of each fault type under different operating conditions, and performing mutual verification; if there is a difference greater than a threshold, marking the corresponding fault type as a fault type to be verified; The correction method of the fault type to be verified is: take the characteristic parameter with the largest prediction probability as the second benchmark, establish a trend analysis curve based on the second benchmark, and the trend analysis curve reflects the changing trend of the characteristic parameter over time or operating conditions; calculate the fault type to be verified of the characteristic parameter and the trend analysis curve, and make corrections based on the trend compliance, including: Assume that there are m fault types in the motor, and j is used to represent the order of the fault types; the predicted probability of the jth fault type in the i-th operating condition is recorded as P ij ; The predicted probability P is calculated by the following formula ij Modified probability of XP ij : ； Among them, xs_ij represents the trend conformity of the characteristic parameters corresponding to the jth fault type in the i-th operating condition; max_xs represents the maximum value of the trend conformity of the fault type; The trend conformity is obtained by taking the characteristic parameter with the largest prediction probability as the second benchmark and establishing a trend analysis curve based on the second benchmark; analyzing the conformity between the characteristic parameter corresponding to the jth fault type in the i-th operating condition and the trend analysis curve to obtain xs_ij.

2. A motor operation fault analysis method based on electric variable analysis according to claim 1, characterized in that: Based on the load conditions and operating status, the operating conditions are obtained, including: Classified by load conditions, motor operating conditions include no-load conditions, rated load conditions, and overload conditions; Classified by operating status, motor operating conditions include starting conditions, braking conditions and stable conditions; Based on the permutations and combinations of load conditions and operating states, 9 operating conditions are obtained.

3. The motor operation fault analysis method based on electric variable analysis according to claim 1 is characterized in that: The motor virtual simulation model is obtained in the following manner: Establish the geometric model of the motor and ensure that the accuracy of the geometric model matches the actual size and shape of the motor; Define material properties: Define material properties and temperature dependencies in the motor; Set boundary conditions and excitation sources: Set boundary conditions and excitation sources according to the actual situation of the motor; set the initial conditions of the simulation; Perform simulation calculation: After setting all parameters and conditions, perform simulation calculation to calculate the performance parameters and working state of the motor; the simulation mode includes steady-state simulation and transient simulation. The steady-state simulation is used to calculate the performance parameters of the motor under a stable working state, and the transient simulation is used to analyze the performance parameters of the motor in a transient process; After the simulation calculation is completed, the simulation results are analyzed and evaluated, and the simulation confidence of each operating condition is recorded.

4. A motor operation fault analysis method based on electric variable analysis according to claim 3, characterized in that: The method for obtaining the fault type probability matrix and the preliminary fault type set includes the following steps: Data collection and organization: Organize the motor operation data of each operating condition, including current, voltage, electric power, power factor, efficiency, iron loss information, torque output, speed, temperature, vibration information and air gap information between rotor and stator; Fault feature library construction: Establish a fault feature library containing characteristic parameters of various fault types under different working conditions. The fault feature library contains detailed descriptions of fault types, characteristic parameters and corresponding threshold range information; Characteristic parameter analysis: The motor operation data of each operating condition is input into the fault diagnosis model, which includes: Feature parameter extraction: extract key feature parameters from the motor operation data of each operating condition; Feature parameter comparison: compare the extracted feature parameters with the data in the fault feature library, and use the similarity to represent the predicted probability of the fault type; Probability judgment: Compare the predicted probability of the fault type with the corresponding preset value. If it exceeds the preset value, it is judged that the corresponding fault type exists; Preliminary fault diagnosis: Based on the results of feature parameter comparison and probability judgment, a preliminary fault type set is obtained, and the fault type probability matrix and preliminary fault type set are output.

5. The method for analyzing motor operation failure based on electric variable analysis according to claim 4, characterized in that: The motor operation fault analysis method includes a step of obtaining a motor fault type set, including: Obtain a fault type probability matrix, and calculate the predicted probability difference or the average predicted probability difference of each fault type under different operating conditions; If the predicted probability difference or the average predicted probability difference of the fault type under different operating conditions is greater than the threshold, it is marked as a fault type to be verified and proceeds to the next step; if the probability difference of the fault type under different operating conditions is not greater than the threshold, the fault type set is not modified and the set A1, A2, ..., A n Take the union and get the motor fault type set QA; The predicted probabilities of the fault types to be verified in the fault type probability matrix are corrected in sequence to obtain a corrected fault type probability matrix; Compare the corrected fault type probability matrix with the corresponding preset value to obtain the corrected fault type set, which is recorded as set XA1, XA2, ..., XA n , and take the union to obtain the corrected motor fault type set QB.

6. A motor operation fault analysis method based on electric variable analysis according to claim 5, characterized in that: When the fault type occurs in only two operating conditions, the predicted probability difference is directly calculated; when the fault type occurs in more than two operating conditions, the maximum predicted probability is used as the first benchmark, and the difference between the remaining predicted probabilities and the first benchmark is calculated to obtain the average probability difference.

7. The motor operation fault analysis method based on electric variable analysis according to claim 1 is characterized in that: Analyze the abnormal information of each operating condition to obtain the abnormal operation coefficient and the growth rate of the abnormal operation coefficient; When the operation abnormality coefficient and the operation abnormality coefficient growth rate exceed the corresponding threshold, the sampling frequency of the next time period is increased.

8. The method for analyzing motor operation failure based on electric variable analysis according to claim 7, characterized in that: The method for obtaining the operation abnormality coefficient is as follows: According to the time occupied by the working condition in the overall operation of the motor, the matching weight coefficient of each working condition is determined to obtain the operating time T of the i-th operating condition i , through the formula Calculate and obtain the weight coefficient of the i-th operating condition; The simulation parameters and actual operating parameters of each operating condition are obtained. Assume that there are R operating parameters in each operating condition, and k represents the sequential number of the operating parameters. The abnormal operating parameters of the i-th condition are calculated by the following formula: ； in, represents the actual parameter of the kth operating parameter of the ith operating condition, represents the simulation prediction parameter of the kth operating parameter of the ith operating condition, represents the rated parameter of the kth operating parameter in the ith operating condition; Based on the abnormal operating parameters of each operating condition, the joint analysis is performed through the formula The operation abnormality coefficient Ya is calculated.

Images