A fault diagnosis method of a permanent magnet synchronous motor applied to a wind power generation system

Abstract

The application provides a method for diagnosing faults of a permanent magnet synchronous motor by collecting multiple source data of a rotor surface temperature value Tzz, a stator coil temperature value Dz, a stator coil resistance value, a stator coil voltage value I, a bearing vibration frequency and a bearing rotating speed of the permanent magnet synchronous motor, improving calculation accuracy, respectively calculating a magnetic coefficient Tcd, a stator coil thermal short circuit coefficient Dxs and a bearing wear vibration coefficient Zdxs, diagnosing faults of the permanent magnet synchronous motor through evaluation and analysis, comprehensively analyzing and calculating a fault diagnosis identification coefficient Sbx, comprehensively analyzing key fault parameters of the permanent magnet synchronous motor, and significantly improving the accuracy of fault diagnosis; through analysis and evaluation of the fault diagnosis identification coefficient Sbx, the unsupervised learning model is finally substituted; the application provides an efficient and reliable solution for deep learning of the permanent magnet synchronous motor, operation monitoring and maintenance, and has important engineering application value and popularization significance.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet synchronous motor fault diagnosis technology, and in particular to a method for diagnosing permanent magnet synchronous motor faults applied to wind power generation systems. Background Technology

[0002] Wind power, as a clean energy technology, plays a vital role in the global energy transition. Permanent magnet synchronous motors (PMSMs) are widely used in wind power systems, especially in direct-drive wind turbine generators, due to their high efficiency, high power density, excellent speed regulation performance, and reliability. However, PMSMs can fail during operation due to various reasons (such as mechanical wear, electrical faults, and environmental factors), affecting the normal operation of wind turbine generators and even causing shutdowns, resulting in economic losses.

[0003] Traditional motor fault diagnosis methods primarily rely on the analysis of single physical quantities, such as temperature, current, and resistance. While these methods can identify faults to some extent, the complexity of motor operating environments and the diversity of fault types mean that the analysis of single physical quantities often falls short of the high-precision fault diagnosis requirements under complex operating conditions. Furthermore, traditional methods depend on human experience to analyze monitoring data, lacking a systematic and intelligent processing approach. This makes the diagnostic results susceptible to human factors, resulting in insufficient accuracy and robustness.

[0004] Based on this, a fault diagnosis method for permanent magnet synchronous motors (PMSMs) applied to wind power generation systems is proposed. An unsupervised learning model is used to train the fault diagnosis method for PSMs in wind power generation systems. This automatically learns the fault diagnosis method, enabling it to capture important feature information in complex nonlinear relationships, significantly improving the accuracy and intelligence level of fault diagnosis. Summary of the Invention

[0005] The purpose of this invention is to provide a fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems, so as to solve the problems mentioned in the background art.

[0006] This invention is achieved through the following technical solution: A fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems, the method comprising the following steps: Step S1: Collect the rotor surface temperature Tzz and stator coil temperature Dz of the permanent magnet synchronous motor using a thermal imager, and collect the stator coil resistance value using a voltmeter and a ohmmeter. The bearing vibration frequency is acquired using an oscilloscope and a vibration sensor, along with the stator coil voltage value I. and bearing speed And build a dataset; Step S2: By extracting the rotor surface temperature value Tzz from the dataset, the rotor demagnetization coefficient Tcd is calculated and obtained. The rotor demagnetization fault is evaluated and analyzed to generate the first screening result of the permanent magnet synchronous motor fault diagnosis, which constitutes the first qualified set of permanent magnet synchronous motors. Step S3: Extract the temperature value Dz and resistance value of the stator coil from the first qualified set of the permanent magnet synchronous motor dataset. The thermal short-circuit coefficient Dxs of the stator winding is calculated based on the voltage value I of the stator winding, and the short-circuit overheating fault of the stator winding is evaluated and analyzed to generate the second screening result of the fault diagnosis of the permanent magnet synchronous motor, which constitutes the second qualified set of the permanent magnet synchronous motor. Step S4: Extract the bearing vibration frequency of the second qualified set of permanent magnet synchronous motors from the dataset. and bearing speed The bearing wear vibration coefficient Zdxs is calculated and obtained, and its abnormal bearing wear vibration fault is evaluated and analyzed to generate the third screening result of permanent magnet synchronous motor fault diagnosis, which constitutes the third qualified set of permanent magnet synchronous motor. Step S5: By correlating the rotor demagnetization coefficient Tcd, the stator coil thermal short-circuit coefficient Dxs, and the bearing wear vibration coefficient Zdxs, the fault diagnosis identification coefficient Sbx is obtained through comprehensive calculation. Step S6: Based on the comprehensive evaluation fault diagnosis identification coefficient Sbx, use a neural network model to diagnose the faults of the permanent magnet synchronous motor and output the correct diagnosis results.

[0007] Specifically, step S2 includes: By extracting the rotor surface temperature value Tzz from the dataset and performing dimensionless processing, the rotor demagnetization coefficient Tcd is calculated using the following formula:

[0008]

[0009] In the formula, This represents the temperature value at the i-th monitoring point on the rotor surface, and n represents the total number of monitoring points. This represents the average value of the rotor surface temperature. This indicates the highest temperature value on the rotor surface. This indicates the demagnetization parameters.

[0010] Specifically, step S2 further includes setting a first standard threshold Q and comparing it with the rotor demagnetization coefficient Tcd, as follows: When the rotor demagnetization coefficient Tcd > the first standard threshold Q, it indicates a rotor demagnetization fault, triggers the first early warning command, adds a first-level fault label, and generates the first screening result for permanent magnet synchronous motor fault diagnosis. When the rotor demagnetization coefficient Tcd ≤ the first standard threshold Q, it indicates that there is no rotor demagnetization fault, triggering the second early warning command, marking the next level of qualified screening, and forming the first qualified set of permanent magnet synchronous motors.

[0011] Specifically, step S3 includes: extracting the temperature value Dz of the stator coil from the first qualified set of the permanent magnet synchronous motor dataset, and after dimensionless processing, calculating the thermal short-circuit coefficient Dxs of the stator coil, the formula of which is as follows:

[0012] In the formula, This represents the temperature value at the j-th monitoring point of the stator coil, where m represents the total number of stator coils. This represents the average temperature value of the stator coil. This indicates the highest temperature value of the stator coil. This indicates the lowest temperature value of the stator coil. This indicates the stator coil voltage value. This indicates the resistance value of the stator coil. This indicates the cooling parameters of the stator coils.

[0013] Specifically, step S3 further includes setting a second standard threshold P and comparing it with the thermal short-circuit coefficient Dxs of the stator coil, as follows: When the thermal short-circuit coefficient Dxs of the stator coil is greater than the second standard threshold P, it indicates a short-circuit overheating fault in the stator winding, triggering the third early warning command, marking a secondary fault label, and generating the second screening result for fault diagnosis of permanent magnet synchronous motor. When the thermal short-circuit coefficient Dxs of the stator coil is less than or equal to the second standard threshold P, it indicates that there is no stator winding short-circuit overheating fault, triggering the fourth early warning instruction, marking the second-level screening qualified label, and forming the second qualified set of permanent magnet synchronous motors.

[0014] Specifically, step S4 includes: extracting the bearing vibration frequency of the second qualified set of the permanent magnet synchronous motor dataset. and bearing speed After dimensionless processing, the bearing wear vibration coefficient Zdxs is calculated using the following formula:

[0015] In the formula, This represents the bearing rotation speed collected for the uth time. Let u represent the bearing vibration frequency collected in the u-th sample, and b represent the total number of samples collected. and All are weighting coefficients.

[0016] Specifically, step S4 further includes setting a third standard threshold K and comparing it with the bearing wear vibration coefficient Zdxs, as follows: When the bearing wear vibration coefficient Zdxs > the third standard threshold K, it indicates an abnormal bearing wear vibration fault, triggers the fifth early warning instruction, marks the third-level fault label, and generates the third screening result for permanent magnet synchronous motor fault diagnosis. When the bearing wear vibration coefficient Zdxs ≤ the third standard threshold K, it indicates that there is no abnormal bearing wear vibration fault, triggering the sixth warning instruction, marking the third-level screening qualified label, and forming the third qualified set of permanent magnet synchronous motor.

[0017] Specifically, step S5 includes calculating the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx by comprehensively calculating the rotor demagnetization coefficient Tcd, stator coil thermal short-circuit coefficient Dxs, and bearing wear vibration coefficient Zdxs after dimensionless processing. The formula is as follows:

[0018] In the formula, , and All are weighting coefficients.

[0019] Specifically, step S6 includes: setting a preset fourth standard threshold G and comparing it with the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx, as follows: When the fault diagnosis identification coefficient Sbx of the permanent magnet synchronous motor is greater than the fourth standard threshold G, it indicates that the permanent magnet synchronous motor is faulty and an abnormal fault diagnosis result is generated.

[0020] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: (1) The fault diagnosis method of permanent magnet synchronous motor applied to wind power generation system combines multi-source data monitoring of permanent magnet synchronous motor with fault screening process. It completes fault screening and classification step by step from multi-dimensional fault types such as rotor demagnetization, stator winding short circuit overheating to bearing wear and vibration, and realizes the standardization and systematization of the diagnosis process.

[0021] (2) The fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems uses equipment such as thermal imagers, oscilloscopes, vibration sensors, voltmeters and ohmmeters to collect data on various physical quantities, including rotor surface temperature, stator coil temperature, stator voltage and resistance, bearing vibration frequency and speed. By calculating the rotor demagnetization coefficient Tcd, stator thermal short-circuit coefficient Dxs, and bearing wear vibration coefficient Zdxs, the key fault parameters of the permanent magnet synchronous motor are comprehensively analyzed, which significantly improves the accuracy of fault diagnosis.

[0022] (3) The fault diagnosis method of permanent magnet synchronous motor applied to wind power generation system generates a first qualified set, a second qualified set and a third qualified set in sequence through a step-by-step screening method. It can effectively classify and locate the fault type and its severity, provide efficient technical support for fault management, and reduce the misdiagnosis rate and missed diagnosis rate.

[0023] (4) This method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems calculates fault diagnosis identification coefficients comprehensively, establishes and trains an unsupervised learning model using neural network technology, trains the fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems, automatically learns the fault diagnosis method, and outputs correct diagnosis results. It is applicable to fault diagnosis of permanent magnet synchronous motors under various complex operating conditions and has wider applicability and reliability. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort. Figure 1 The flowchart of a fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0026] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0027] It should be understood that the invention can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0028] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising” and / or “including,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0029] To fully understand this invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by this invention. Optional embodiments of the invention are described in detail below; however, in addition to these detailed descriptions, the invention may have other embodiments.

[0030] Please see Figure 1 A fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems, the method comprising the following steps: Step S1: Collect the rotor surface temperature Tzz and stator coil temperature Dz of the permanent magnet synchronous motor using a thermal imager, and collect the stator coil resistance value using a voltmeter and a ohmmeter. The bearing vibration frequency is acquired using an oscilloscope and a vibration sensor, along with the stator coil voltage value I. and bearing speed And build a dataset; Step S2: By extracting the rotor surface temperature value Tzz from the dataset, the rotor demagnetization coefficient Tcd is calculated and obtained. The rotor demagnetization fault is evaluated and analyzed to generate the first screening result of the permanent magnet synchronous motor fault diagnosis, which constitutes the first qualified set of permanent magnet synchronous motors. Step S3: Extract the temperature value Dz and resistance value of the stator coil from the first qualified set of the permanent magnet synchronous motor dataset. The thermal short-circuit coefficient Dxs of the stator winding is calculated based on the voltage value I of the stator winding, and the short-circuit overheating fault of the stator winding is evaluated and analyzed to generate the second screening result of the fault diagnosis of the permanent magnet synchronous motor, which constitutes the second qualified set of the permanent magnet synchronous motor. Step S4: Extract the bearing vibration frequency of the second qualified set of permanent magnet synchronous motors from the dataset. and bearing speed The bearing wear vibration coefficient Zdxs is calculated and obtained, and its abnormal bearing wear vibration fault is evaluated and analyzed to generate the third screening result of permanent magnet synchronous motor fault diagnosis, which constitutes the third qualified set of permanent magnet synchronous motor. Step S5: By correlating the rotor demagnetization coefficient Tcd, the stator coil thermal short-circuit coefficient Dxs, and the bearing wear vibration coefficient Zdxs, the fault diagnosis identification coefficient Sbx is obtained through comprehensive calculation. Step S6: Based on the comprehensive evaluation fault diagnosis identification coefficient Sbx, use a neural network model to diagnose the faults of the permanent magnet synchronous motor and output the correct diagnosis results.

[0031] In this embodiment, the permanent magnet synchronous motor is positioned on a fault diagnosis and testing platform. A thermal imager, oscilloscope, vibration sensor, voltmeter, and ohmmeter are used to monitor the rotor surface temperature, stator coil temperature, stator coil voltage and resistance, bearing vibration frequency, and bearing speed of the permanent magnet synchronous motor in real time. The rotor surface temperature value Tzz, stator coil temperature value Dz, and stator coil resistance value are collected. The voltage value I of the stator coil and the bearing vibration frequency. and bearing speed Multi-source data is used to improve computational accuracy. The magnetic coefficient Tcd, stator coil thermal short-circuit coefficient Dxs, and bearing wear vibration coefficient Zdxs are calculated separately. Through evaluation and analysis, faults in permanent magnet synchronous motors (PMSMs) are diagnosed. Then, a comprehensive analysis is performed to obtain the fault diagnosis identification coefficient Sbx, and a comprehensive analysis of key fault parameters of PMSMs significantly improves the accuracy of fault diagnosis. The fault diagnosis identification coefficient Sbx is analyzed and evaluated, and finally substituted into an unsupervised learning model. This provides an efficient and reliable solution for fault diagnosis, operation monitoring, and maintenance of deep learning-based PMSMs, possessing significant engineering application value and promotional significance.

[0032] Specifically, step S2 includes: By extracting the rotor surface temperature value Tzz from the dataset and performing dimensionless processing, the rotor demagnetization coefficient Tcd is calculated. This calculation provides a basis for effectively improving the accuracy of fault identification. The formula is as follows:

[0033]

[0034] In the formula, This represents the temperature value at the i-th monitoring point on the rotor surface, and n represents the total number of monitoring points. This represents the average value of the rotor surface temperature. This indicates the highest temperature value on the rotor surface. This indicates the demagnetization parameter, with commonly used values ​​ranging from -0.10% / ℃ to 0.13% / ℃.

[0035] Specifically, step S2 further includes setting a first standard threshold Q and comparing it with the rotor demagnetization coefficient Tcd, as follows: When the rotor demagnetization coefficient Tcd > the first standard threshold Q, it indicates a rotor demagnetization fault, triggers the first early warning command, adds a first-level fault label, and generates the first screening result for permanent magnet synchronous motor fault diagnosis. When the rotor demagnetization coefficient Tcd ≤ the first standard threshold Q, it indicates that there is no rotor demagnetization fault, triggering the second early warning command, marking the next level of qualified screening, and forming the first qualified set of permanent magnet synchronous motors.

[0036] Specifically, step S3 includes: extracting the temperature value Dz of the stator coil from the first qualified set of the permanent magnet synchronous motor dataset, and after dimensionless processing, calculating the thermal short-circuit coefficient Dxs of the stator coil, the formula of which is as follows:

[0037] In the formula, This represents the temperature value at the j-th monitoring point of the stator coil, where m represents the total number of stator coils. This represents the average temperature value of the stator coil. This indicates the highest temperature value of the stator coil. This indicates the lowest temperature value of the stator coil. This indicates the stator coil voltage value. This indicates the resistance value of the stator coil. This indicates the cooling parameters of the stator coil, with commonly used values ​​ranging from 30 to 100 W / ℃.

[0038] In this embodiment, the thermal short-circuit coefficient Dxs of the stator coil is calculated by dimensionlessly processing the temperature value Dz of the stator coil in the first qualified set of the permanent magnet synchronous motor. This effectively improves the ability to identify complex motor faults and meets the high-precision diagnosis requirements of permanent magnet synchronous motors under complex operating conditions.

[0039] Specifically, step S3 further includes setting a second standard threshold P and comparing it with the thermal short-circuit coefficient Dxs of the stator coil, as follows: When the thermal short-circuit coefficient Dxs of the stator coil is greater than the second standard threshold P, it indicates a short-circuit overheating fault in the stator winding, triggering the third early warning command, marking a secondary fault label, and generating the second screening result for fault diagnosis of permanent magnet synchronous motor. When the thermal short-circuit coefficient Dxs of the stator coil is less than or equal to the second standard threshold P, it indicates that there is no stator winding short-circuit overheating fault, triggering the fourth early warning instruction, marking the second-level screening qualified label, and forming the second qualified set of permanent magnet synchronous motors.

[0040] In this embodiment, by comparing and analyzing the thermal short-circuit coefficient Dxs of the stator coil, the presence of a stator winding short-circuit overheating fault in the permanent magnet synchronous motor is evaluated. After evaluation, a secondary screening of the fault is performed, which effectively ensures the high efficiency of fault diagnosis.

[0041] Specifically, step S4 includes: extracting the bearing vibration frequency of the second qualified set of the permanent magnet synchronous motor dataset. and bearing speed After dimensionless processing, the bearing wear vibration coefficient Zdxs is calculated using the following formula:

[0042] In the formula, This represents the bearing rotation speed collected for the uth time. Let u represent the bearing vibration frequency collected in the u-th sample, and b represent the total number of samples collected. and All are weighting coefficients. , ,and .

[0043] In this embodiment, the bearing vibration frequency of the second qualified set of the permanent magnet synchronous motor is used. and bearing speed The calculation of the bearing wear vibration coefficient Zdxs provides data support for judging abnormal bearing wear vibration faults in permanent magnet synchronous motors, significantly improving the accuracy of fault diagnosis. Specifically, step S4 further includes setting a third standard threshold K and comparing it with the bearing wear vibration coefficient Zdxs, as follows: When the bearing wear vibration coefficient Zdxs > the third standard threshold K, it indicates an abnormal bearing wear vibration fault, triggers the fifth early warning instruction, marks the third-level fault label, and generates the third screening result for permanent magnet synchronous motor fault diagnosis. When the bearing wear vibration coefficient Zdxs ≤ the third standard threshold K, it indicates that there is no abnormal bearing wear vibration fault, triggering the sixth warning instruction, marking the third-level screening qualified label, and forming the third qualified set of permanent magnet synchronous motor.

[0044] In this embodiment, by analyzing and evaluating the bearing wear vibration coefficient Zdxs, it is determined whether there is an abnormal bearing wear vibration fault in the permanent magnet synchronous motor. After evaluation, a three-level screening of faults is performed, providing efficient technical support for fault management.

[0045] Specifically, step S5 includes calculating the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx by comprehensively calculating the rotor demagnetization coefficient Tcd, stator coil thermal short-circuit coefficient Dxs, and bearing wear vibration coefficient Zdxs after dimensionless processing. The formula is as follows:

[0046] In the formula, , and All are weighting coefficients. , , ,and .

[0047] Specifically, step S6 includes: setting a preset fourth standard threshold G and comparing it with the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx, as follows: When the fault diagnosis identification coefficient Sbx of the permanent magnet synchronous motor is greater than the fourth standard threshold G, it indicates that the permanent magnet synchronous motor is faulty and an abnormal fault diagnosis result is generated.

[0048] A fault diagnosis identification coefficient Sbx ≤ fourth standard threshold G for permanent magnet synchronous motors indicates that the permanent magnet synchronous motor has no faults, and a fault-free diagnosis result for permanent magnet synchronous motors is generated. In this embodiment, in the preceding steps, the rotor demagnetization coefficient (Tcd), stator coil thermal short-circuit coefficient (Dxs), and bearing wear vibration coefficient (Zdxs) were obtained by extracting various fault features, and the fault diagnosis identification coefficient (Sbx) was obtained by correlating these features.

[0049] These coefficients, used as input data, will be employed in the neural network model for judgment. This neural network model comprises an input layer, hidden layers, and an output layer. The input layer receives the fault identification coefficients (Sbx). The hidden layers propagate forward through each layer of the network, with each node processing the received information according to an activation function (such as ReLU, Sigmoid, etc.), passing the information layer by layer until the output layer. The output layer provides the fault diagnosis result.

[0050] In this embodiment, by comprehensively calculating the fault diagnosis and identification coefficient Sbx, this method supports diverse diagnostic results outputs such as fault trend analysis, fault early warning prompts, and fault category identification, which facilitates further development of maintenance strategies and provides comprehensive protection for the reliability and stability of motor operation.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fault diagnosis method for permanent magnet synchronous motors applied to wind power generation systems, characterized in that, The method includes the following steps: Step S1: Collect the rotor surface temperature Tzz and stator coil temperature Dz of the permanent magnet synchronous motor using a thermal imager, and collect the stator coil resistance value using a voltmeter and a ohmmeter. The bearing vibration frequency is acquired using an oscilloscope and a vibration sensor, along with the stator coil voltage value I. and bearing speed And build a dataset; Step S2: By extracting the rotor surface temperature value Tzz from the dataset, and after dimensionless processing, the rotor demagnetization coefficient Tcd is calculated and obtained. The rotor demagnetization fault is then evaluated and analyzed to generate the first screening result for fault diagnosis of permanent magnet synchronous motor, forming the first qualified set of permanent magnet synchronous motors. The specific formula for calculating the rotor demagnetization coefficient Tcd is as follows: In the formula, This represents the temperature value at the i-th monitoring point on the rotor surface, and n represents the total number of monitoring points. This represents the average value of the rotor surface temperature. This indicates the highest temperature value on the rotor surface. Indicates demagnetization parameters; Step S3: Extract the temperature value Dz and resistance value of the stator coil from the first qualified set of the permanent magnet synchronous motor dataset. The voltage value I of the stator coil is dimensionlessly processed to calculate the thermal short-circuit coefficient Dxs of the stator coil. This is then used to evaluate and analyze the short-circuit overheating fault of the stator winding, generating a second screening result for permanent magnet synchronous motor fault diagnosis. This constitutes the second qualified set of permanent magnet synchronous motors. The specific formula for calculating the thermal short-circuit coefficient Dxs of the stator coil is as follows: In the formula, This represents the temperature value at the j-th monitoring point of the stator coil, where m represents the total number of stator coils. This represents the average temperature value of the stator coil. This indicates the highest temperature value of the stator coil. This indicates the lowest temperature value of the stator coil. This indicates the stator coil voltage value. This indicates the resistance value of the stator coil. These represent the cooling parameters of the stator coils; Step S4: Extract the bearing vibration frequency of the second qualified set of permanent magnet synchronous motors from the dataset. and bearing speed After dimensionless processing, the bearing wear vibration coefficient Zdxs is calculated and obtained. An evaluation and analysis of abnormal bearing wear vibration faults is then performed, generating the third screening result for permanent magnet synchronous motor fault diagnosis, forming the third qualified set of permanent magnet synchronous motors. The specific formula for calculating the bearing wear vibration coefficient Zdxs is as follows: In the formula, This represents the bearing rotation speed collected for the uth time. Let u represent the bearing vibration frequency collected in the u-th sample, and b represent the total number of samples collected. and All are weighting coefficients; Step S5: By correlating the rotor demagnetization coefficient Tcd calculated in step S2, the stator coil thermal short-circuit coefficient Dxs calculated in step S3, and the bearing wear vibration coefficient Zdxs calculated in step S4, the fault diagnosis identification coefficient Sbx is obtained through comprehensive calculation. Step S6: Based on the comprehensive evaluation fault diagnosis identification coefficient Sbx, use a neural network model to diagnose the faults of the permanent magnet synchronous motor and output the correct diagnosis results.

2. The method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems according to claim 1, characterized in that, Step S2 further includes setting a first standard threshold Q and comparing it with the rotor demagnetization coefficient Tcd, as follows: When the rotor demagnetization coefficient Tcd > the first standard threshold Q, it indicates a rotor demagnetization fault, triggers the first early warning command, adds a first-level fault label, and generates the first screening result for permanent magnet synchronous motor fault diagnosis. When the rotor demagnetization coefficient Tcd ≤ the first standard threshold Q, it indicates that there is no rotor demagnetization fault, triggering the second early warning command, marking the next level of qualified screening, and forming the first qualified set of permanent magnet synchronous motors.

3. The method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems according to claim 1, characterized in that, Step S3 further includes setting a second standard threshold P and comparing it with the thermal short-circuit coefficient Dxs of the stator coil, as detailed below: When the thermal short-circuit coefficient Dxs of the stator coil is greater than the second standard threshold P, it indicates a short-circuit overheating fault in the stator winding, triggering the third early warning command, marking a secondary fault label, and generating the second screening result for fault diagnosis of permanent magnet synchronous motor. When the thermal short-circuit coefficient Dxs of the stator coil is less than or equal to the second standard threshold P, it indicates that there is no stator winding short-circuit overheating fault, triggering the fourth early warning instruction, marking the second-level screening qualified label, and forming the second qualified set of permanent magnet synchronous motors.

4. The method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems according to claim 1, characterized in that, Step S4 further includes setting a third standard threshold K and comparing it with the bearing wear vibration coefficient Zdxs, as follows: When the bearing wear vibration coefficient Zdxs > the third standard threshold K, it indicates an abnormal bearing wear vibration fault, triggers the fifth early warning instruction, marks the third-level fault label, and generates the third screening result for permanent magnet synchronous motor fault diagnosis. When the bearing wear vibration coefficient Zdxs ≤ the third standard threshold K, it indicates that there is no abnormal bearing wear vibration fault, triggering the sixth warning instruction, marking the third-level screening qualified label, and forming the third qualified set of permanent magnet synchronous motor.

5. The method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems according to claim 1, characterized in that, Step S5 specifically includes calculating the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx by taking the rotor demagnetization coefficient Tcd, the stator coil thermal short-circuit coefficient Dxs, and the bearing wear vibration coefficient Zdxs, after dimensionless processing, as follows: In the formula, , and All are weighting coefficients.

6. The method for fault diagnosis of permanent magnet synchronous motors applied to wind power generation systems according to claim 1, characterized in that, Step S6 can also be: A preset fourth standard threshold G is used, and a comparative analysis is performed with the permanent magnet synchronous motor fault diagnosis identification coefficient Sbx, as detailed below: When the fault diagnosis identification coefficient Sbx of the permanent magnet synchronous motor is greater than the fourth standard threshold G, it indicates that the permanent magnet synchronous motor is faulty and an abnormal fault diagnosis result is generated.

Images