Performance evaluation method and system for permanent magnet synchronous motor based on intelligent perception

Abstract

The invention discloses a permanent magnet synchronous motor performance evaluation method and system based on intelligent perception, and relates to the field of motor engineering and control, and the method comprises the steps: collecting the operation data and historical operation data of a permanent magnet synchronous motor in real time, carrying out the preprocessing, building a fault mode classification model based on the preprocessed motor operation data, and carrying out the calculation of the fault mode classification model. And outputting a motor fault result and a health index, establishing a motor performance prediction model according to the motor fault result and the health index, and outputting a motor performance prediction result. According to the method, the time domain and frequency domain features are combined, and the Euclidean distance calculation is carried out, so that the accuracy of fault mode recognition is improved, faults can be found and positioned in time, the health index can be calculated, the health condition of the motor can be comprehensively evaluated, and an important basis is provided for subsequent performance prediction. Besides, the motor performance prediction model is established, the periodic change rule of the health index is captured through nonlinear fitting, and the long-term descending trend is combined, so that the accuracy of performance prediction is improved, and the service life of the motor is prolonged.

Description

Technical Field

[0001] The present invention relates to the field of motor engineering and control, and in particular to a permanent magnet synchronous motor performance evaluation method and system based on intelligent perception. Background Art

[0002] Permanent magnet synchronous motors have been widely used in industrial automation, electric vehicles, wind power generation, etc. due to their high efficiency, high power density and good dynamic response characteristics. Traditional motor performance evaluation methods mainly rely on regular manual inspection and maintenance, which is inefficient and difficult to detect potential faults in time.

[0003] The existing motor performance evaluation methods have improved the monitoring and evaluation capabilities of the motor operating status to a certain extent, but there are still some shortcomings. The existing fault mode classification models are usually based on static features and lack comprehensive consideration of time domain and frequency domain features, resulting in limited accuracy of fault identification. In addition, most of the existing motor performance prediction models rely on linear regression or simple nonlinear fitting, which fails to fully capture the long-term trend and periodic change law of the motor health index, and the accuracy of the prediction results needs to be improved. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a permanent magnet synchronous motor performance evaluation method based on intelligent perception to solve the problem that the existing fault mode classification model is usually based on static features and lacks comprehensive consideration of time domain and frequency domain features, resulting in limited accuracy of fault identification.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception, which comprises collecting the operating data and historical operating data of the permanent magnet synchronous motor in real time and performing preprocessing; Establish a fault mode classification model based on the preprocessed motor operation data, and output the motor fault results and health index; According to the motor fault results and health index, a motor performance prediction model is established to output the motor performance prediction results; Based on the motor operation data, the dynamic parameters of the permanent magnet synchronous motor are identified online using the least squares method; The motor operation state model is used to calculate the online identification results to obtain the motor dynamic parameters; The current performance index of the motor is calculated based on the motor dynamic parameters and the motor performance prediction results.

[0007] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, the operation data and historical operation data of the permanent magnet synchronous motor are collected in real time and preprocessed, which specifically includes the following steps: Real-time collection of three-phase current, three-phase voltage and speed of permanent magnet synchronous motor during operation; Collecting historical operation data of the permanent magnet synchronous motor, wherein the historical operation data includes three-phase current, three-phase voltage, speed, fault type, fault occurrence time and fault mode; Use a low-pass filter to remove noise from real-time motor operation data; Use linear interpolation to fill in missing values ​​in real-time motor operation data; Use statistical methods to identify and remove outliers in real-time motor operation data.

[0008] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, wherein: a fault mode classification model is established based on pre-processed historical operation data, and the motor fault results and health index are output, which specifically includes the following steps: Obtaining a fault mode feature template and a normal state template; The time series signal of the motor operation data is divided into multiple sliding windows to extract time domain features; Use fast Fourier transform to convert the motor operation data from the time domain to the frequency domain and extract the spectrum characteristics of the motor operation data; Combine the features of time domain and frequency domain to form a fault classification feature matrix; The distance between the fault classification feature matrix and the fault mode feature template is calculated by Euclidean distance, and its expression is: ; in, Represents the fault classification feature matrix and the The distance of the class failure mode template, Indicates the first eigenvalues, Indicates Class Failure Mode eigenvalues, Indicates the total number of features; Determine the current motor operation fault mode according to the minimum distance between the fault classification feature matrix and the fault mode feature template; According to the degree of deviation between the fault classification feature matrix and the normal state template, the current motor health index is calculated, and its expression is: ; in, Indicates the current health index of the motor. Indicates The weight of the feature value, Indicates The template value of a feature value in the normal state.

[0009] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, wherein: according to the fault mode classification result and the health index, a motor performance prediction model is established, and the motor performance prediction result is output, which specifically includes the following steps: Extract health index time series from historical operation data; Using the historical health index time series, the long-term downward trend of the health index is fitted; Calculate the influence value of the current fault mode according to the current motor operation fault mode and the weight in the corresponding fault mode feature template; Nonlinear fitting is used to capture the periodic change law of the health index in the historical operation data, and the long-term downward trend of the health index and the impact value of the current motor operation fault mode are combined to output the motor performance prediction value, which is expressed as: ; in, Indicated in Time-predicted motor performance prediction values, Indicates the health index when the motor is fault-free and performance is not degraded. represents the number of historical health indices, Indicates The contribution of each historical time point to the current forecast, represents the frequency parameter, express Historical health index at the moment, Indicates the decline in health index. Indicates the speed at which the health index declines. Indicates the current time point. represents the initial time point, Indicates the impact of the current motor operation fault mode on the health index.

[0010] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, the dynamic parameters of the permanent magnet synchronous motor are identified online based on the motor operation data using the least squares method, which specifically includes the following steps: Set the initial parameters of stator resistance and permanent magnet flux according to the historical operation data of the motor, and define the initial covariance matrix according to the initial parameters; Set the error threshold based on the motor's historical operating data; The gain vector is calculated based on the real-time data of the motor and the covariance matrix of the previous moment, and its expression is: ; in, Indicates at time The gain vector at Indicated in The covariance matrix at time , Indicates at time The three-phase current value, Indicates three-phase current value The transpose of represents the forgetting factor; The online identification result of the dynamic parameters is obtained by calculating the gain vector, the real-time motor data and the dynamic parameter vector at the previous moment: ; in, Indicates at time The dynamic parameter vector of Indicates at time The dynamic parameter vector at time , Indicates at time The three-phase voltage value, Indicates three-phase current value The transpose of When the error between the dynamic parameter vector and the true parameter vector is less than the error threshold, the online identification result of the motor dynamic parameters is obtained.

[0011] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, wherein: using the motor operation state model to calculate the online identification result to obtain the motor dynamic parameters, specifically includes the following steps: Based on the online identification results of the motor dynamic parameters, numerical integration is used and the gamma function is introduced for calculation to obtain the motor dynamic parameters, which are expressed as follows: ; in, represents the final calculated motor dynamic parameters, represents the attenuation factor, represents the time in the integration interval, Indicates at time The dynamic parameter vector of represents the integral differential, () represents the gamma function, Represents the order of the fractional integral.

[0012] As a preferred solution of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in the present invention, wherein: according to the motor dynamic parameters and the motor performance prediction results, the current performance index of the motor is calculated, which specifically includes the following steps: The motor dynamic parameters and the motor health index are weightedly integrated to form a comprehensive parameter vector; The motor performance prediction value is corrected according to the comprehensive vector parameters, and the current performance index of the motor is calculated. The expression is: ; in, Indicates at time Motor performance indicators, represents the corrected motor performance prediction value, represents the mean of all motor performance prediction values ​​before correction, represents the standard deviation of all motor performance prediction values ​​before correction, Represents the smoothing factor.

[0013] In a second aspect, the present invention provides a permanent magnet synchronous motor performance evaluation system based on intelligent perception, including a data acquisition and processing module, which collects the operation data and historical operation data of the permanent magnet synchronous motor in real time and performs preprocessing; The fault classification module builds a fault mode classification model based on the preprocessed motor operation data and outputs the motor fault results and health index; The motor prediction module establishes a motor performance prediction model based on the motor fault results and health index, and outputs the motor performance prediction results; Online identification module, based on motor operation data, uses the least squares method to perform online identification of the dynamic parameters of the permanent magnet synchronous motor; The dynamic parameter calculation module uses the motor operation state model to calculate the online identification results to obtain the motor dynamic parameters; The performance index module calculates the current performance index of the motor based on the motor dynamic parameters and the motor performance prediction results.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the permanent magnet synchronous motor performance evaluation method based on intelligent perception as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the permanent magnet synchronous motor performance evaluation method based on intelligent perception as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are: combining time domain and frequency domain features, through Euclidean distance calculation, the accuracy of fault mode recognition is improved, which helps to timely discover and locate faults, calculate the health index, and comprehensively evaluate the health status of the motor, providing an important basis for subsequent performance prediction. In addition, a motor performance prediction model is established, and the periodic change law of the health index is captured through nonlinear fitting, and combined with the long-term downward trend, the accuracy of performance prediction is improved. Accurate performance prediction can help discover potential problems in advance, provide a basis for preventive maintenance, and extend the life of the motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a flow chart of the permanent magnet synchronous motor performance evaluation method based on intelligent perception in Example 1.

[0019] Figure 2 This is a schematic diagram of the failure mode classification model in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a permanent magnet synchronous motor performance evaluation method based on intelligent perception, comprising the following steps: S1. Collect the operating data and historical operating data of the permanent magnet synchronous motor in real time and perform preprocessing.

[0024] The specific steps include: S1.1. Use high-precision sensors to collect the three-phase current, three-phase voltage and speed of the permanent magnet synchronous motor in real time at a frequency of 10kHz to ensure high time resolution monitoring of the dynamic characteristics of the motor operation; In addition to real-time data, S1.2 collects historical operating data of the permanent magnet synchronous motor, including three-phase current, three-phase voltage, speed, fault type, fault occurrence time and fault mode.

[0025] S1.3, select the low-pass filter and set the cut-off frequency to 1kHz to remove high-frequency noise; Through data integrity checks, missing data points are identified. After filling in the missing values, the data becomes more continuous, avoiding analysis errors caused by missing data; Use the Z-score method to identify outliers. For example, set the threshold to 3. When the absolute value of the Z-score of a data point is greater than 3, the data point is considered an outlier and the identified outlier is removed from the data set to avoid its impact on data analysis and model training.

[0026] It is further explained that after removing outliers, the data is purer, the quality of the data is improved, the impact of outliers on data analysis results is reduced, and the accuracy of the analysis is improved.

[0027] S2. Establish a fault mode classification model based on the preprocessed historical operation data, and output the fault mode classification results and health index.

[0028] The specific steps include: S2.1. Extract the features of different fault modes (such as stator winding short circuit, rotor eccentricity, etc.) from the historical operation data. According to the extracted features, establish a feature template for each fault mode. At the same time, establish a normal state feature template for comparison with the motor operation state under fault-free conditions.

[0029] S2.2, dividing the time series signal of the motor operation data into multiple sliding windows, and extracting the mean, variance, skewness and kurtosis in each window; It is further explained that the extracted features are mainly used to describe the global characteristics of the signal (such as central value, energy, shape, etc.), providing a basis for subsequent classification.

[0030] The motor operation data is converted from the time domain to the frequency domain using fast Fourier transform, and the spectrum characteristics including the main frequency component and spectrum energy are extracted.

[0031] It is further explained that the frequency domain features are used to describe the periodicity and energy distribution of the signal, and can reflect the vibration characteristics related to faults in the operation state of the motor.

[0032] S2.3. After extracting the time domain and frequency domain features, all features are combined into a fault classification feature matrix , used for fault classification.

[0033] ; It is further explained that by combining the time domain and frequency domain features, the motor operation status can be analyzed from multiple dimensions to improve the accuracy of fault diagnosis. And the form of feature matrix is ​​convenient for subsequent calculation and processing.

[0034] S2.4. Use the Euclidean distance to calculate the distance between the current feature vector and the fault mode feature. The expression is: ; in, Represents the fault classification feature matrix and the The distance of the class failure mode template, Indicates the first eigenvalues, Indicates Class Failure Mode eigenvalues, Indicates the total number of features; At the same time, the distance between the current feature vector and the normal state template is calculated.

[0035] Whether the current motor is faulty and what the motor operation fault mode is is determined based on the minimum Euclidean distance between the fault classification feature matrix and the fault mode feature template and the normal state template.

[0036] S2.5. Calculate the current motor health index based on the degree of deviation between the fault classification feature matrix and the normal state template. The expression is: ; in, Indicates the current health index of the motor. Indicates The weight of the feature value, Indicates The template value of a feature value in the normal state.

[0037] To further explain, the health index provides a quantitative indicator for evaluating the overall health of the motor.

[0038] S3. According to the fault mode classification results and health index, a motor performance prediction model is established to output the motor performance prediction results.

[0039] The specific steps include: S3.1 Constants derived from statistical analysis of historical health indices .

[0040] Extracting health index time series from historical operation data ; Using the historical health index time series, the long-term downward trend of the health index is fitted; The weights in the current motor operation fault mode and the corresponding fault mode feature template are calculated, and the indicator function is introduced to obtain the impact value of the current fault mode on the health index. , whose expression is: ; in, Indicates the impact of the current fault mode on the health index. Indicates the total number of fault types. Indicates Class failure mode, Indicates The impact weight of the class failure mode on the health index, represents the failure mode classification result, Indicates the indicator function, indicating whether the current motor is in the If yes, the value is 1, otherwise it is 0.

[0041] It is further explained that different fault modes have different effects on the health index, and this adjustment helps to more accurately reflect the actual condition of the motor.

[0042] Nonlinear fitting is used to capture the periodic change law of the health index in the historical operation data, and the long-term downward trend of the health index and the impact value of the current motor operation fault mode are combined to output the motor performance prediction value, which is expressed as: ; in, Indicated in Time-predicted motor performance prediction values, Indicates the health index when the motor is fault-free and performance is not degraded. represents the number of historical health indices, Indicates The contribution of each historical time point to the current forecast, represents the frequency parameter, express Historical health index at the moment, Indicates the decline in health index. Indicates the speed at which the health index declines. Indicates the current time point. represents the initial time point, Indicates the impact of the current motor operation fault mode on the health index.

[0043] It is further illustrated that the effects of long-term trends, cyclical variations, and current fault modes are combined to provide a comprehensive motor performance prediction.

[0044] S4. Based on the motor operation data, the dynamic parameters of the permanent magnet synchronous motor are identified online using the least squares method.

[0045] The specific steps include: S4.1. Calculate and set initial parameters of stator resistance and permanent magnet flux linkage according to the historical operation data of the motor, and define an initial covariance matrix according to the initial parameters. The parameters of stator resistance and permanent magnet flux linkage are dynamic parameters.

[0046] The covariance matrix reflects the uncertainty of the parameters and is set as a diagonal matrix whose diagonal elements represent the initial uncertainty of each parameter.

[0047] It is further explained that calculating and setting the initial parameters through historical operation data provides a reasonable starting point and helps to speed up the convergence.

[0048] S4.2. According to the parameter variation range and measurement noise level in the historical operation data, a reasonable error threshold is set. This threshold is used to determine whether the parameter estimation converges to the true value.

[0049] S4.3. Calculate the gain vector based on the real-time data of the motor and the covariance matrix of the previous moment, and its expression is: ; in, Indicates at time The gain vector at Indicated in The covariance matrix at time , Indicates at time The three-phase current value, Indicates three-phase current value The transpose of represents the forgetting factor; It is further explained that the gain vector enables the dynamic parameters to be adaptively adjusted to better reflect the current operating state of the motor.

[0050] S4.4. Calculate based on the gain vector and the dynamic parameter vector of the previous moment to obtain the online identification result of the dynamic parameter: ; in, Indicates at time The dynamic parameter vector of Indicates at time The dynamic parameter vector at time , Indicates at time The three-phase voltage value, Indicates three-phase current value The transpose of It is further explained that, through the recursive formula, the dynamic parameters can be updated in real time to promptly reflect the current state of the motor. As data continues to accumulate, the accuracy of dynamic parameter estimation will gradually improve and become closer to the true value.

[0051] S4.5. The calculated dynamic parameter vector Compare with the true parameter vector.

[0052] If the error is less than the preset error threshold, it is considered that the parameter estimation has converged and the final dynamic parameter identification result is obtained.

[0053] S5. Use the motor operation state model to calculate the online identification results to obtain the motor dynamic parameters.

[0054] The specific steps include: Set the time range of the integration, determine the attenuation factor and the order of the fractional integration; The time range is divided into multiple small segments, the integral of each small segment is calculated, and the gamma function is introduced for calculation to obtain the dynamic parameters of the motor, which are expressed as follows: ; in, represents the final calculated motor dynamic parameters, represents the attenuation factor, represents the time in the integration interval, Indicates at time The dynamic parameter vector of represents the integral differential, () represents the gamma function, Represents the order of the fractional integral.

[0055] It is further shown that the introduction of fractional-order integral helps to improve the stability of dynamic parameter estimation and reduce the impact of noise and outliers.

[0056] S6. Calculate the current performance index of the motor according to the motor dynamic parameters and the motor performance prediction results, which specifically includes the following steps: The motor dynamic parameters and the motor health index are weighted and fused to form a comprehensive parameter vector, which is expressed as: ; in, Indicates at time The comprehensive parameter vector when Indicates at time The motor health index at the time.

[0057] The motor performance prediction value is corrected according to the comprehensive vector parameters, and its expression is: ; in, represents the corrected motor performance prediction value, represents the dimension of the comprehensive vector parameter, Indicates The weight of the comprehensive vector parameter, Indicates at time No. The value of the integrated vector parameter, Indicates the initial state The value of a comprehensive vector parameter.

[0058] The current performance index of the motor is calculated based on the corrected motor performance prediction value, and its expression is: ; in, Indicates at time Motor performance indicators, represents the corrected motor performance prediction value, represents the mean of all motor performance prediction values ​​before correction, represents the standard deviation of all motor performance prediction values ​​before correction, Represents the smoothing factor.

[0059] The value range of the motor's current performance index is divided into positive values, negative values ​​and zero values.

[0060] when When it is greater than zero and is a positive value, it means that the motor's performance is better than the average level. The larger the positive value, the better the motor's performance.

[0061] when When it is less than zero and is a negative value, it means that the performance of the motor is below average. The larger the negative value, the worse the performance of the motor.

[0062] when When it is equal to zero, it means that the performance of the motor is at an average level.

[0063] This embodiment also provides a permanent magnet synchronous motor performance evaluation system based on intelligent perception, including: Data acquisition and processing module, which collects the operating data and historical operating data of the permanent magnet synchronous motor in real time and performs preprocessing; The fault classification module builds a fault mode classification model based on the preprocessed motor operation data and outputs the motor fault results and health index; The motor prediction module establishes a motor performance prediction model based on the motor fault results and health index, and outputs the motor performance prediction results; Online identification module, based on motor operation data, uses the least squares method to perform online identification of the dynamic parameters of the permanent magnet synchronous motor; The dynamic parameter calculation module uses the motor operation state model to calculate the online identification results to obtain the motor dynamic parameters; The performance index module calculates the current performance index of the motor based on the motor dynamic parameters and the motor performance prediction results.

[0064] This embodiment also provides a computer device, which is suitable for the permanent magnet synchronous motor performance evaluation method based on intelligent perception, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the permanent magnet synchronous motor performance evaluation method based on intelligent perception proposed in the above embodiment.

[0065] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0066] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (Static Random Access Memory, referred to as SRAM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), an erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), a programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), a read-only memory (Read-Only Memory, referred to as ROM), a magnetic storage device, a flash memory, a disk or an optical disk.

[0067] In summary, the present invention improves the accuracy of fault mode recognition by combining time domain and frequency domain features and calculating the Euclidean distance, which helps to timely discover and locate faults, calculate the health index, and comprehensively evaluate the health status of the motor, providing an important basis for subsequent performance prediction. In addition, a motor performance prediction model is established, and the periodic change law of the health index is captured through nonlinear fitting, and combined with the long-term downward trend, the accuracy of performance prediction is improved. Accurate performance prediction can help discover potential problems in advance, provide a basis for preventive maintenance, and extend the life of the motor.

[0068] Example 2, referring to Table 1, Table 2, Table 3 and Table 4, is the second embodiment of the present invention. In order to further verify the technical solution of the present invention, experimental simulation data of a permanent magnet synchronous motor performance evaluation method based on intelligent perception are provided.

[0069] In order to verify the effectiveness and advantages of the permanent magnet synchronous motor performance evaluation method based on intelligent perception, the following experiment was designed. The experimental object is a permanent magnet synchronous motor (PMSM) with a rated power of 5kW, which is used in industrial automation production lines. The experiment is divided into two stages: the first stage is data acquisition and preprocessing, and the second stage is fault mode classification.

[0070] Phase 1: Data collection and preprocessing Real-time data collection: High-precision sensors are used to collect the three-phase current, three-phase voltage and speed of the permanent magnet synchronous motor in real time. The sampling frequency is set to 10kHz to ensure that subtle changes in the motor's operating status are captured.

[0071] The details are shown in Table 1: Table 1: Permanent magnet synchronous motor data sheet Collect historical operation data of permanent magnet synchronous motor, including three-phase current, three-phase voltage, speed, fault type, fault occurrence time and fault mode. The time span of historical data is 6 months to cover the operation under different working conditions.

[0072] Data preprocessing: Noise removal: A low-pass filter is used to denoise the three-phase current and voltage signals collected in real time. The cutoff frequency of the filter is set to 1kHz to effectively remove high-frequency noise.

[0073] Missing value filling: For missing data caused by sensor failure or communication interruption, linear interpolation is used to fill in the missing data to ensure data continuity and integrity.

[0074] Outlier detection: Use the Z-score method to identify and remove outliers. Set the threshold to 3, that is, when the absolute value of the Z-score of a data point is greater than 3, the data point is considered an outlier and is removed.

[0075] Phase 2: Failure Mode Classification Failure mode classification: The features of different fault modes are extracted from the historical operation data. Based on the extracted features, a feature template is established for each fault mode. At the same time, a feature template of the normal state is also established to compare the motor operation state under the fault-free condition.

[0076] The details are shown in Table 2: Table 2: Motor fault mode feature template and normal state template The real-time collected motor operation data is divided into multiple sliding windows, each with a length of 1 second. The time domain features of each window are extracted, such as mean, variance, skewness, and peak value.

[0077] Use Fast Fourier Transform (FFT) to convert the time domain signal to the frequency domain and extract the spectrum characteristics, such as the main frequency component and spectrum energy.

[0078] Combine the time domain and frequency domain features to form a fault classification feature matrix; The details are shown in Table 3: Table 3: Motor fault classification characteristics The distance between the fault classification feature matrix and the fault mode feature template is calculated through the Euclidean distance to determine the current motor operation fault mode.

[0079] Taking motor B as an example, the expression for calculating distance is: ; The distance to the stator winding short circuit is: ; The distance from the rotor eccentricity is: ; The distance from the normal template is: ; The current motor operation fault mode is determined based on the minimum distance between the fault classification feature matrix and the fault mode feature template, so the fault mode of motor B is rotor eccentricity.

[0080] The other motors have obtained the current motor operation fault mode using the above calculation method, as shown in Table 4: Table 4: Motor current operation fault mode table Through the above steps and data tables, the operating status of the permanent magnet synchronous motor is evaluated by calculating the Euclidean distance between the time domain and frequency domain of the current motor operating data and the fault template and the normal template.

[0081] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A permanent magnet synchronous motor performance evaluation method based on intelligent perception, characterized in that: include, Collect the operating data and historical operating data of the permanent magnet synchronous motor in real time and perform preprocessing; Establish a fault mode classification model based on the preprocessed motor operation data, and output the motor fault results and health index; According to the motor fault results and health index, a motor performance prediction model is established to output the motor performance prediction results; Based on the motor operation data, the dynamic parameters of the permanent magnet synchronous motor are identified online using the least squares method; The motor operation state model is used to calculate the online identification results to obtain the motor dynamic parameters; The current performance index of the motor is calculated based on the motor dynamic parameters and the motor performance prediction results.

2. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 1, characterized in that: Real-time collection of permanent magnet synchronous motor operation data and historical operation data, and pre-processing, specifically including the following steps: Real-time collection of three-phase current, three-phase voltage and speed of permanent magnet synchronous motor during operation; Collecting historical operation data of the permanent magnet synchronous motor, wherein the historical operation data includes three-phase current, three-phase voltage, speed, fault type, fault occurrence time and fault mode; Use a low-pass filter to remove noise from real-time motor operation data; Use linear interpolation to fill in missing values ​​in real-time motor operation data; Use statistical methods to identify and remove outliers in real-time motor operation data.

3. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 2, characterized in that: Based on the pre-processed historical operation data, a fault mode classification model is established to output the motor fault results and health index. The specific steps include: Establish failure mode feature templates and normal state templates based on historical operation data; The time series signal of the motor operation data is divided into multiple sliding windows to extract time domain features; Use fast Fourier transform to convert the motor operation data from the time domain to the frequency domain and extract the spectrum characteristics of the motor operation data; Combine the features of time domain and frequency domain to form a fault classification feature matrix; The distance between the fault classification feature matrix and the fault mode feature template is calculated by Euclidean distance, and its expression is: ； in, Represents the fault classification feature matrix and the The distance of the class failure mode template, Indicates the first eigenvalues, Indicates Class Failure Mode eigenvalues, Indicates the total number of features; Determine the current motor operation fault mode according to the minimum distance between the fault classification feature matrix and the fault mode feature template; According to the degree of deviation between the fault classification feature matrix and the normal state template, the current motor health index is calculated, and its expression is: ； in, Indicates the current health index of the motor. Indicates The weight of the feature value, Indicates The template value of a feature value in the normal state.

4. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 3, characterized in that: According to the fault mode classification results and health index, a motor performance prediction model is established to output the motor performance prediction results. The specific steps include: Extract health index time series from historical operation data; Using the historical health index time series, the long-term downward trend of the health index is fitted; Calculate the influence value of the current fault mode according to the current motor operation fault mode and the weight in the corresponding fault mode feature template; Nonlinear fitting is used to capture the periodic change law of the health index in the historical operation data, and the long-term downward trend of the health index and the impact value of the current motor operation fault mode are combined to output the motor performance prediction value, which is expressed as: ； in, Indicated in Time-predicted motor performance prediction values, Indicates the health index when the motor is fault-free and performance is not degraded. represents the number of historical health indices, Indicates The contribution of each historical time point to the current forecast, represents the frequency parameter, express Historical health index at the moment, Indicates the decline in health index. Indicates the speed at which the health index declines. Indicates the current time point. represents the initial time point, Indicates the impact of the current motor operation fault mode on the health index.

5. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 4, characterized in that: Based on the motor operation data, the least squares method is used to perform online identification of the dynamic parameters of the permanent magnet synchronous motor, which specifically includes the following steps: The initial parameters of the stator resistance and permanent magnet flux are set according to the historical operation data of the motor, and the initial covariance matrix is ​​defined according to the initial parameters; Set the error threshold based on the motor's historical operating data; The gain vector is calculated based on the real-time data of the motor and the covariance matrix of the previous moment, and its expression is: ； in, Indicates at time The gain vector at Indicated in The covariance matrix at time , Indicates at time The three-phase current value, Indicates three-phase current value The transpose of represents the forgetting factor; The online identification result of the dynamic parameters is obtained by calculating the gain vector, the real-time motor data and the dynamic parameter vector at the previous moment: ； in, Indicates at time The dynamic parameter vector of Indicates at time The dynamic parameter vector at time , Indicates at time The three-phase voltage value, Indicates three-phase current value The transpose of When the error between the dynamic parameter vector and the true parameter vector is less than the error threshold, the online identification result of the motor dynamic parameters is obtained.

6. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 5, characterized in that: The motor operation state model is used to calculate the online identification results to obtain the motor dynamic parameters, which specifically includes the following steps: Based on the online identification results of the motor dynamic parameters, numerical integration is used and the gamma function is introduced for calculation to obtain the motor dynamic parameters, which are expressed as follows: ； in, represents the final calculated motor dynamic parameters, represents the attenuation factor, represents the time in the integration interval, Indicates at time The dynamic parameter vector of represents the integral differential, () represents the gamma function, Represents the order of the fractional integral.

7. The method for evaluating the performance of a permanent magnet synchronous motor based on intelligent perception according to claim 6, characterized in that: According to the motor dynamic parameters and motor performance prediction results, the current performance indicators of the motor are calculated, which specifically includes the following steps: The motor dynamic parameters and the motor health index are weightedly integrated to form a comprehensive parameter vector; The motor performance prediction value is corrected according to the comprehensive vector parameters, and the current performance index of the motor is calculated. The expression is: ； in, Indicates at time Motor performance indicators, represents the corrected motor performance prediction value, represents the mean of all motor performance prediction values ​​before correction, represents the standard deviation of all motor performance prediction values ​​before correction, Represents the smoothing factor.

8. A permanent magnet synchronous motor performance evaluation system based on intelligent perception, based on the permanent magnet synchronous motor performance evaluation method and system based on intelligent perception according to any one of claims 1 to 7, characterized in that: include, Data acquisition and processing module, which collects the operating data and historical operating data of the permanent magnet synchronous motor in real time and performs preprocessing; The fault classification module builds a fault mode classification model based on the preprocessed motor operation data and outputs the motor fault results and health index; The motor prediction module establishes a motor performance prediction model based on the motor fault results and health index, and outputs the motor performance prediction results; Online identification module, based on motor operation data, uses the least squares method to perform online identification of the dynamic parameters of the permanent magnet synchronous motor; The dynamic parameter calculation module uses the motor operation state model to calculate the online identification results to obtain the motor dynamic parameters; The performance index module calculates the current performance index of the motor based on the motor dynamic parameters and the motor performance prediction results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the permanent magnet synchronous motor performance evaluation method based on intelligent perception described in any one of claims 1 to 7 are implemented.

Images