A method for detecting defects of motor magnetic steel sheets based on vibration analysis

Abstract

The present application provides a method for detecting defects of motor magnetic steel sheets based on vibration analysis, which relates to the technical field of motor fault diagnosis, and includes: extracting vibration characteristics from multiple vibration data sets, respectively inputting the multiple vibration characteristic sets into a defect characteristic recognition library for similarity traversal comparison, and outputting multiple initial defect characteristics; collecting the temperature distribution characteristics and acoustic wave characteristics of a target motor within a predetermined period, performing a credible probability analysis on the multiple initial defect characteristics, and setting the initial defect characteristic with the highest probability as the detection result of the defects of the motor magnetic steel sheets. Through the present application, the technical problem that the traditional method for detecting defects of motor magnetic steel sheets cannot quickly and accurately identify the defect characteristics of motor magnetic steel sheets during the operation of the motor, resulting in insufficient timeliness and accuracy of defect detection, can be solved. Through the detection method based on vibration analysis and multi-characteristic fusion, the efficiency, accuracy and reliability of detecting defects of motor magnetic steel sheets can be significantly improved.

Description

Technical Field

[0001] This application relates to the technical field of motor fault diagnosis, and particularly to a method for detecting defects in motor magnetic steel sheets based on vibration analysis. Background Art

[0002] As an important power equipment in industrial production, the health status of a motor directly affects the operation efficiency, production safety, and maintenance cost of the equipment.

[0003] As an important component of a motor, the motor magnetic steel sheet is under high-frequency magnetic fields and mechanical stresses for a long time, and is prone to defects such as cracks, detachment, and wear. These defects will cause unbalanced operation, increased vibration, and decreased efficiency of the motor, and even lead to major failures, resulting in shutdowns and equipment damage. Therefore, timely and accurate detection of defects in motor magnetic steel sheets is crucial for ensuring the safe operation of the motor. Summary of the Invention

[0004] The purpose of this application is to provide a method for detecting defects in motor magnetic steel sheets based on vibration analysis, so as to solve the technical problem that the traditional method for detecting defects in motor magnetic steel sheets cannot quickly and accurately identify the defect characteristics of motor magnetic steel sheets during motor operation, resulting in insufficient timeliness and accuracy of defect detection.

[0005] In view of the above problems, this application provides a method for detecting defects in motor magnetic steel sheets based on vibration analysis, including: building a vibration monitoring array based on predetermined monitoring points, continuously monitoring a target motor through the vibration monitoring array, and obtaining multiple vibration data sets within a predetermined period; extracting vibration characteristics from the multiple vibration data sets to obtain multiple vibration feature sets; respectively inputting the multiple vibration feature sets into a defect feature recognition library for similarity traversal comparison, selecting multiple defect feature sets that meet the dynamic comparison threshold, and performing intersection analysis on the multiple defect feature sets to output multiple initial defect features; collecting the temperature distribution characteristics and acoustic wave characteristics of the target motor within a predetermined period, and performing credible probability analysis on the multiple initial defect features according to the temperature distribution characteristics and acoustic wave characteristics, and outputting the initial defect feature with the highest probability as the detection result of the motor magnetic steel sheet defect of the target motor.

[0006] The technical solution provided in this application has at least the following technical effects or advantages:

[0007] By building a vibration monitoring array based on predetermined monitoring points, continuously monitoring the target motor through the vibration monitoring array, and obtaining multiple vibration data sets within a predetermined period; then extracting vibration characteristics from the multiple vibration data sets to obtain multiple vibration characteristic sets; further inputting the multiple vibration characteristic sets into a defect characteristic recognition library respectively for similarity traversal comparison, selecting multiple defect characteristic sets that meet the dynamic comparison threshold, and performing intersection analysis on the multiple defect characteristic sets to output multiple initial defect characteristics; on the other hand, collecting the temperature distribution characteristics and acoustic wave characteristics of the target motor within a predetermined period; then performing credible probability analysis on the multiple initial defect characteristics according to the temperature distribution characteristics and acoustic wave characteristics, and outputting the initial defect characteristic with the highest probability as the defect detection result of the motor magnetic steel sheet of the target motor; that is to say, through multi-point vibration monitoring similarity comparison, multiple relevant potential defect characteristics can be quickly discovered, and then through the joint analysis and credibility discrimination of the temperature and acoustic wave characteristics on the multiple relevant potential defect characteristics, the defect characteristic with the highest credibility is determined, thereby significantly improving the efficiency, accuracy and reliability of the defect detection of the motor magnetic steel sheet.

[0008] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specifically illustrates the specific embodiments of the present application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0010] Figure 1 It is a schematic flow chart of a method for detecting defects in motor magnetic steel sheets based on vibration analysis according to the present application.

[0011] Figure 2 It is a schematic flow chart of obtaining multiple vibration characteristic sets in a method for detecting defects in motor magnetic steel sheets based on vibration analysis according to the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] The present application provides a method for detecting defects in motor magnetic steel sheets based on vibration analysis, which solves the technical problems that the traditional method for detecting defects in motor magnetic steel sheets cannot quickly and accurately identify the defect characteristics of motor magnetic steel sheets during the operation of the motor, resulting in insufficient timeliness and accuracy of defect detection. Through multi-point vibration monitoring and similarity comparison, multiple relevant potential defect characteristics can be quickly discovered. Then, through the joint analysis and credibility discrimination of temperature and acoustic wave characteristics for multiple relevant potential defect characteristics, the defect characteristic with the highest credibility is determined, thereby significantly improving the efficiency, accuracy, and reliability of defect detection for motor magnetic steel sheets.

[0013] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the drawings rather than all of them.

[0014] Embodiment, please refer to the attached Figure 1 , the present application provides a method for detecting defects in motor magnetic steel sheets based on vibration analysis, which specifically includes the following steps:

[0015] S100: Build a vibration monitoring array based on predetermined monitoring points, continuously monitor the target motor through the vibration monitoring array, and obtain multiple vibration data sets within a predetermined period.

[0016] Furthermore, step S100 of the present application further includes:

[0017] S110: Obtain the predetermined monitoring points of the target motor, where the predetermined monitoring points at least include the motor stator, the motor rotor, and the motor bearing; S120: Respectively arrange vibration sensors at the motor stator, the motor rotor, and the motor bearing to construct a vibration monitoring array; S130: Continuously monitor the target motor through the vibration monitoring array according to the predetermined monitoring frequency, and obtain multiple vibration data sets of multiple monitoring points within a predetermined period.

[0018] Specifically, first, obtain the predetermined monitoring points of the target motor, where the predetermined monitoring points at least include the motor stator, the motor rotor, and the motor bearings. These monitoring points cover the core components of the motor and can effectively reflect the overall operating state of the motor and the relevant signals of the magnet steel sheet defects. Then, vibration sensors are respectively arranged at the motor stator, the motor rotor, and the motor bearings, and each vibration sensor is connected to a data acquisition system to construct a comprehensive vibration monitoring array. This array coordinates the work of each sensor to ensure that the vibration signals of each key part can be collected and analyzed in real time. Then, according to the operating characteristics of the motor, set an appropriate monitoring frequency (such as collecting a certain number of vibration data per second) to ensure that all vibration changes during the operation of the motor can be captured, especially the weak vibration signals related to the magnet steel sheet defects. Finally, according to the predetermined monitoring frequency, continuously monitor the target motor through the vibration monitoring array to obtain multiple vibration data sets of multiple monitoring points (motor stator, motor rotor, and motor bearings) within a predetermined period (such as within the next 10 minutes). By obtaining multiple vibration data sets of multiple monitoring points, it provides basic data support for subsequent defect analysis.

[0019] S200: Extract vibration characteristics from the multiple vibration data sets to obtain multiple vibration characteristic sets.

[0020] Furthermore, as Figure 2 shown, step S200 of this application further includes:

[0021] S210: Analyze the noise points and perform filtering and denoising processing on the multiple vibration data sets to obtain multiple noise point ratios and multiple standard vibration data sets; S220: Configure vibration characteristic indicators, where the vibration characteristic indicators at least include abnormal frequency components, abnormal frequency ratios, amplitude curves, and waveform curves; S230: Extract vibration characteristics from the multiple standard vibration data sets according to the vibration characteristic indicators to obtain multiple vibration characteristic sets of multiple monitoring points.

[0022] Specifically, first, perform noise point analysis and filtering and denoising processing on the multiple vibration data sets respectively. For example, use statistical analysis methods (such as mean and standard deviation analysis) to identify abnormal fluctuations or noise points in the data. Common noise points include abnormal signals caused by sensor failures and external interferences. The higher the noise point ratio, the more noise components there are in the signal. Use appropriate filtering algorithms to filter the vibration data to remove high-frequency noise or low-frequency interference. Common filtering methods include wavelet transform, Kalman filter, mean filter, etc. Obtain multiple noise point ratios and multiple standard vibration data sets, where the noise point ratio reflects the relative influence degree of noise in the data, and the standard vibration data set is the clear signal after denoising processing.

[0023] Next, configure the vibration characteristic indicators. Among them, the vibration characteristic indicators are used for subsequent feature extraction and fault analysis. These characteristic indicators can effectively capture abnormal changes and potential fault signs in the vibration signal, and at least include abnormal frequency components, abnormal frequency ratios, amplitude curves, and waveform curves. Abnormal frequency components are usually related to motor faults (such as cracked or detached magnetic steel sheets, etc.). The frequency components of the vibration signal are extracted through Fourier transform or fast Fourier transform (FFT), and specific abnormal frequency components (such as multiple frequency characteristics of 1X, 2X, 3X, etc.) are identified and compared with the frequency pattern of a normal motor. The abnormal frequency ratio is the ratio of the abnormal frequency component in the entire frequency spectrum, which is used to indicate the severity of the abnormality. The higher the ratio, the greater the possibility of a fault. The amplitude curve can intuitively show the change in the vibration intensity of the motor at different time periods. Abnormal amplitude changes are often related to internal faults or imbalances in the motor. The waveform curve reflects the change of the vibration signal over time. By analyzing the shape of the waveform curve, the fault mode inside the motor can be identified.

[0024] Then, according to the vibration characteristic indicators, vibration feature extraction is performed on the multiple standard vibration data sets. For example, methods such as Fourier transform (FFT) are used to extract the frequency components of the vibration signal, and abnormal frequency components and corresponding amplitudes are identified. The changes in the vibration signal are analyzed through time-domain statistical features (such as peak value, root mean square value, standard deviation, etc.) to identify possible vibration abnormalities. The shape of the waveform of the vibration signal is analyzed to detect whether there are irregular vibration waveforms. Then, by combining time-domain, frequency-domain, and waveform analysis, multi-dimensional vibration characteristics of each monitoring point (stator, rotor, bearing, etc.) are extracted, and multiple vibration feature sets of multiple monitoring points are obtained. Among them, the vibration feature set includes indicators such as abnormal frequency components, abnormal frequency ratios, amplitude curves, and waveform curves. These feature sets provide a detailed basis for subsequent defect identification and diagnosis.

[0025] S300: Input the multiple vibration feature sets into the defect feature recognition library respectively for similarity traversal and comparison, select multiple defect feature sets that meet the dynamic comparison threshold, and perform intersection analysis on the multiple defect feature sets to output multiple initial defect features.

[0026] Furthermore, step S300 of the present application further includes:

[0027] S310: Construct a defect feature recognition library based on the motor attribute features, the motor magnetic steel sheet attribute features, and the vibration characteristic indicators.

[0028] Furthermore, step S310 of the present application further includes:

[0029] S311: Using the motor attribute features and the motor magnet sheet attribute features as equipment constraints, the vibration characteristic indicators as conditional constraints, and the defect detection of the motor magnet sheet based on vibration analysis as a guide, perform information retrieval based on big data to obtain multiple sample vibration characteristic sets, and label the defect types and defect intensities of the motor magnet sheets under different sample vibration characteristic sets to obtain multiple sample defect types and multiple sample defect intensities; S312: Cluster the multiple sample defect types and multiple sample defect intensities based on the same defect type and defect intensity intervals to determine multiple standard defect types, where each standard defect type is marked with multiple defect intensity intervals; S313: Randomly select a first standard defect type and a first defect intensity interval of the first standard defect type, and screen the multiple sample vibration characteristic sets according to the first standard defect type and the first defect intensity interval to determine multiple first sample vibration characteristic sets; S314: Perform high-frequency feature analysis on the multiple first sample vibration characteristic sets to determine a first standard vibration characteristic set, where the first standard vibration characteristic set includes a first standard abnormal frequency component, a first standard abnormal frequency ratio, a first standard amplitude curve, and a first standard waveform curve; S315: Establish a first mapping association between the first standard vibration characteristic set and the first standard defect type and the first defect intensity interval, and construct a first defect recognition branch according to the first mapping association; S316: Analyze and obtain multiple defect recognition branches in sequence, and construct the defect feature recognition library according to the multiple defect recognition branches.

[0030] Specifically, first, obtain the motor attribute features and the motor magnet sheet attribute features. Among them, the motor attribute features include the motor type, operating state, motor component characteristics, etc.; the motor magnet sheet attribute features include material characteristics, magnet sheet shape and size, etc. Then, use the motor attribute features and the motor magnet sheet attribute features as equipment constraints, that is, limit the acquisition range and type of vibration characteristics through equipment constraints, use the vibration characteristic indicators as conditional constraints, and use the defect detection of the motor magnet sheet based on vibration analysis as a guide. Perform information retrieval based on big data, that is, based on the motor operating state, the monitoring data of vibration sensors, and the different working conditions of the motor magnet sheet, collect the vibration data sets of multiple motors through the data acquisition system; and after using the big data analysis platform for data cleaning and preprocessing, extract the vibration characteristics that meet the equipment constraints and conditional constraints to form multiple sample vibration characteristic sets. Further label the defect types and defect intensities of the motor magnet sheets under different sample vibration characteristic sets, that is, perform manual labeling on each vibration characteristic set or classify it through intelligent analysis algorithms (such as machine learning algorithms), and label the defect types (such as cracks, peeling, warping, wear, etc.) of the motor magnet sheets and their corresponding defect intensities (mild, moderate, severe); obtain multiple sample defect types and multiple sample defect intensities.

[0031] Next, cluster the multiple sample defect types and multiple sample defect intensities based on the same defect type and defect intensity interval. For example, use clustering algorithms (such as K-means, DBSCAN, hierarchical clustering, etc.) to perform clustering analysis on the defect types and intensities. The purpose of this step is to group samples with similar characteristics into the same category, thereby identifying multiple standard defect types and their corresponding defect intensity intervals. Clustering analysis maps the relationship between defect types and defect intensities to ensure that defects of the same type are grouped together and different intensity intervals are divided according to the change in defect intensity, determining multiple standard defect types (such as cracks, peeling, warping, wear). Among them, each standard defect type is marked with multiple defect intensity intervals.

[0032] Then, randomly select any one of the multiple standard defect types as the first standard defect type, and randomly select the first defect intensity interval (any one intensity interval) of the first standard defect type. Next, screen the multiple sample vibration feature sets according to the first standard defect type and the first defect intensity interval to determine multiple first sample vibration feature sets. Further perform high-frequency feature analysis on the multiple first sample vibration feature sets. By performing high-frequency feature analysis on the multiple first sample vibration feature sets, high-frequency components related to defects are extracted. These frequency components can usually reflect the nature and intensity of the defects. For example, select the mode value of the multiple first sample vibration feature sets as the first standard vibration feature set to obtain the first standard vibration feature set. Among them, the first standard vibration feature set includes the first standard abnormal frequency component, the first standard abnormal frequency ratio, the first standard amplitude curve, and the first standard waveform curve.

[0033] Further establish the first mapping association between the first standard vibration feature set and the first standard defect type and the first defect intensity interval. This mapping relationship associates different vibration features with the specific type and intensity of the defects, providing a basis for subsequent defect diagnosis. Then, based on the decision tree principle, use the first standard vibration feature set as the child node, and the first standard defect type and the first defect intensity interval as the leaf nodes of the child node. Construct the first defect recognition branch according to the first mapping association. This branch is based on the conditional constraints of the first standard vibration feature set and the defect type intensity and can perform precise defect recognition for specific defect types and intensities. Use the same method to analyze and construct multiple defect recognition branches in turn, and then integrate the multiple defect recognition branches into a complete defect feature recognition library. This library contains different defect types, different defect intensities, and their corresponding vibration features, thus providing a comprehensive basis for the online defect detection of motor magnetic steel sheets.

[0034] S320: Calculate the mean value of the multiple noise ratios to determine the comprehensive noise coefficient, calculate the ratio of the comprehensive noise coefficient to the standard noise coefficient, and obtain the threshold compensation coefficient; S330: Compensate the initial similarity comparison threshold according to the threshold compensation coefficient to obtain the dynamic comparison threshold.

[0035] Specifically, in vibration data, noise points may affect the extraction of vibration characteristics and the accuracy of defect feature recognition. Therefore, it is necessary to analyze the noise points and calculate the comprehensive noise coefficient to evaluate the impact of noise and perform compensation in subsequent processing. First, calculate the mean value of the multiple noise ratios, and set the mean calculation result as the comprehensive noise coefficient. Then, calculate the ratio of the comprehensive noise coefficient to the standard noise coefficient (the ideal noise level obtained based on historical data or theoretical analysis) to obtain the threshold compensation coefficient. The compensation coefficient reflects the difference between the actual noise and the standard noise and can be used to adjust the comparison threshold. Further, compensate the initial similarity comparison threshold according to the threshold compensation coefficient, that is, multiply the initial similarity comparison threshold by the reciprocal of the threshold compensation coefficient. For example, assume the threshold compensation coefficient is 0.95 and the initial similarity comparison threshold is 80%, indicating that the data quality is relatively high. Therefore, it is necessary to increase the similarity comparison threshold. Then, multiply 80% by the reciprocal of 0.95, which is approximately 0.84%, and set it as the dynamic comparison threshold; obtain the dynamic comparison threshold. Among them, the larger the reciprocal of the threshold compensation coefficient, the slightly higher the dynamic comparison threshold, making the defect recognition more stringent and reducing the possibility of false alarms; the smaller the reciprocal of the threshold compensation coefficient, the appropriately lower the dynamic comparison threshold, making the tolerance of defect recognition higher, thereby reducing missed alarms. Through this method of adjusting the dynamic comparison threshold, it is possible to effectively cope with the noise changes during the operation of the motor, improve the flexibility and accuracy of defect detection. When the noise environment is relatively complex, automatically adjusting the comparison threshold ensures that the system can not only ensure accuracy but also avoid missed detections caused by overly strict thresholds; when the data quality is relatively high, appropriately increasing the comparison threshold can more accurately identify potential defects and ensure the accuracy and reliability of the detection results.

[0036] S340: Input the multiple vibration feature sets into the defect feature recognition library respectively for similarity traversal comparison, and select the defect features whose similarity meets the dynamic comparison threshold, and output the multiple defect feature sets, where the defect features include the defect type and defect intensity of the motor magnetic steel sheet, and the defect type includes at least crack, detachment, warping, and wear.

[0037] Furthermore, step S340 of the present application further includes:

[0038] S341: Randomly select the first vibration feature set and randomly select the first defect recognition branch in the defect feature recognition library.

[0039] Specifically, first, randomly select any one of the multiple vibration feature sets and set it as the first vibration feature set, and randomly select a first defect recognition branch from the multiple defect recognition branches of the defect feature recognition library.

[0040] S342: Input the first vibration feature set into the first defect recognition branch, and perform a similarity traversal comparison with the first standard vibration feature set. If the comparison similarities all meet the dynamic comparison threshold, then output the first standard defect type and the first defect intensity range and set them as the first defect feature.

[0041] Furthermore, step S342 of the present application further includes:

[0042] S3421: Obtain the first vibration feature set, where the first vibration feature set includes a first abnormal frequency component, a first abnormal frequency ratio, a first amplitude curve, and a first waveform curve; S3422: Perform a similarity comparison between the first abnormal frequency component and the first standard abnormal frequency component to determine a first similarity; S3423: If the first similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first abnormal frequency ratio and the first standard abnormal frequency ratio to determine a second similarity; S3424: If the second similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first amplitude curve and the first standard amplitude curve to determine a third similarity; S3425: If the third similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first waveform curve and the first standard waveform curve to determine a fourth similarity. If the fourth similarity is greater than the dynamic comparison threshold, then output the first standard defect type and the first defect intensity range and set them as the first defect feature.

[0043] Specifically, obtain the first vibration feature set, where the first vibration feature set includes a first abnormal frequency component, a first abnormal frequency ratio, a first amplitude curve, and a first waveform curve; then, perform a similarity comparison between the first abnormal frequency component and the first standard abnormal frequency component, such as by calculating the Euclidean distance for similarity comparison, to obtain a first similarity; if the first similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first abnormal frequency ratio and the first standard abnormal frequency ratio to determine a second similarity; if the second similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first amplitude curve and the first standard amplitude curve to determine a third similarity; if the third similarity is greater than the dynamic comparison threshold, then perform a similarity comparison between the first waveform curve and the first standard waveform curve to determine a fourth similarity, and if the fourth similarity is greater than the dynamic comparison threshold, then set the output of the first standard defect type (such as crack, shedding, wear, etc.) and the first defect intensity interval as the first defect feature. Wherein, if the first similarity is less than or equal to the dynamic comparison threshold or the second similarity is less than or equal to the dynamic comparison threshold or the third similarity is less than or equal to the dynamic comparison threshold or the fourth similarity is less than or equal to the dynamic comparison threshold, then stop this comparison.

[0044] S343: Sequentially perform a similarity traversal comparison between the first vibration feature set and the standard vibration feature sets of other defect recognition branches in the defect feature recognition library, obtain a first defect feature set, and add it to the multiple defect feature sets.

[0045] Specifically, using the same method, sequentially perform a similarity traversal comparison between the first vibration feature set and the standard vibration feature sets of other defect recognition branches in the defect feature recognition library, output a first defect feature set that meets the dynamic comparison threshold, and sequentially obtain multiple defect feature sets of multiple monitoring points. Finally, perform an intersection analysis on the multiple defect feature sets and output multiple initial defect features. Through the intersection analysis, the defect features jointly supported among multiple monitoring points and different feature sets can be found, noise and redundant features can be removed, and the most representative defect features can be extracted, significantly improving the accuracy, reliability, and robustness of the defect detection of the motor magnetic steel sheet.

[0046] S400: Collect the temperature distribution characteristics and acoustic wave characteristics of the target motor within a predetermined period, perform a credible probability analysis on the multiple initial defect features according to the temperature distribution characteristics and acoustic wave characteristics, and set the initial defect feature with the highest probability as the defect detection result of the motor magnetic steel sheet of the target motor.

[0047] Furthermore, step S400 of the present application further includes:

[0048] S410: Using the motor attribute features and the motor magnet sheet attribute features as device constraints, the defect type and defect intensity as feature constraints, and the motor magnet sheet defect detection as a guide, retrieve and obtain a sample temperature distribution feature set, a sample acoustic wave feature set, and multiple sample defect feature sets, and count the proportion of sample defect features under different sample temperature distributions and different sample acoustic wave features, which is set as the trigger probability of multiple defect features, construct a sample defect feature probability distribution, and obtain a sample defect feature probability distribution set; S420: Use the sample temperature distribution feature set, the sample acoustic wave feature set, the multiple sample defect feature sets, and the sample defect feature probability distribution set to perform supervised training on N prediction operators until convergence, to obtain N defect feature trigger probability prediction units, and integrally construct a defect feature trigger probability predictor based on the N defect feature trigger probability prediction units, where the prediction operators at least include a BP neural network, a random forest, and a support vector machine, and N is an integer greater than or equal to 3; S430: Input the temperature distribution feature, the acoustic wave feature, and the multiple initial defect features into the defect feature trigger probability predictor for credible probability analysis, output a defect feature probability distribution, and select the initial defect feature with the highest probability in the defect feature probability distribution as the motor magnet sheet defect detection result of the target motor.

[0049] Specifically, in the defect detection of motor magnet sheets, in addition to vibration data, temperature distribution features and acoustic wave features are also very important auxiliary information. By combining these features, the accuracy and reliability of defect detection can be further improved. First, collect the temperature distribution features and acoustic wave features of the target motor within a predetermined period. For example, within a predetermined period (such as every hour or every minute), collect the temperature data of the target motor at each monitoring point to generate a temperature distribution feature set. Temperature changes may be related to defects in the motor magnet sheet (such as thermal expansion, friction, etc.), so temperature features help detect the presence of defects; deploy acoustic wave sensors outside the motor or at other appropriate positions to collect the acoustic wave signals generated during the operation of the motor. Defects in the motor magnet sheet are often accompanied by abnormal sound changes, such as collisions, friction, or vibrations, which will all be reflected in the acoustic wave signals; obtain the acoustic wave data of the target motor during operation through the acoustic wave sensors to generate an acoustic wave feature set. The acoustic wave features can be extracted by analyzing parameters such as the acoustic wave spectrum, intensity, and waveform.

[0050] Next, taking the motor attribute characteristics and the motor magnet sheet attribute characteristics as device constraints, the defect type and defect intensity as feature constraints, and the motor magnet sheet defect detection as a guide, a sample temperature distribution feature set, a sample acoustic wave feature set, and multiple sample defect feature sets are obtained based on big data retrieval; further, the proportion of sample defect features under different sample temperature distributions and different sample acoustic wave features is statistically calculated, and the feature proportion is set as the triggering probability of the defect feature, obtaining the triggering probabilities of multiple defect features. Combining multiple sample defect features to construct a sample defect feature probability distribution, a sample defect feature probability distribution set is obtained.

[0051] Configure N prediction operators, where the prediction operators at least include BP neural network, random forest, and support vector machine. N is an integer greater than or equal to 3, and the value of N can be set according to the actual situation; then the sample temperature distribution feature set, the sample acoustic wave feature set, multiple sample defect feature sets, and the sample defect feature probability distribution set are used as sample training data, and the sample training data is equally divided into N parts. Select N times with replacement from the N parts of data to construct a first training set, and select N times iteratively to obtain N training sets. Then, taking the sample temperature distribution feature, the sample acoustic wave feature, and multiple sample defect features as inputs, and the sample defect feature probability distribution as the output, the BP neural network, random forest, and support vector machine are respectively supervised and trained through the N training sets until convergence, obtaining N defect feature triggering probability prediction units; then the N defect feature triggering probability prediction units are verified through a validation set, obtaining the N prediction accuracies of the N defect feature triggering probability prediction units. Further, configure N output weights according to the N prediction accuracies, where the weight is positively correlated with the prediction accuracy, that is, the greater the prediction accuracy, the greater the weight. Finally, the N defect feature triggering probability prediction units are integrated and fused according to the N output weights to construct a defect feature triggering probability predictor, where the output of the defect feature triggering probability predictor is the weighted calculation result of the N defect feature triggering probability prediction units. Through this ensemble learning method, the advantages of multiple prediction operators can be fully utilized, and the prediction accuracy can be improved through weighted fusion. The output results of different prediction operators are assigned different weights according to their performance on the validation set, so that in practical applications, the final defect prediction result is more accurate and robust, and can handle the motor magnet sheet defect detection task under complex working conditions.

[0052] Then, the temperature distribution feature, the acoustic wave feature, and multiple initial defect features are input into the defect feature triggering probability predictor for credible probability analysis, and a defect feature probability distribution is output; finally, the initial defect feature with the maximum probability in the defect feature probability distribution is selected as the motor magnet sheet defect detection result of the target motor. Through this solution, the defect type and intensity of the motor magnet sheet can be identified in real time and efficiently, significantly improving the accuracy and reliability of the motor magnet sheet defect detection.

[0053] In summary, the method for detecting defects in motor magnetic steel sheets based on vibration analysis provided by the present application has the following technical effects:

[0054] By building a vibration monitoring array based on predetermined monitoring points, continuously monitoring the target motor through the vibration monitoring array, and obtaining multiple vibration data sets within a predetermined period; then extracting vibration characteristics from the multiple vibration data sets to obtain multiple vibration characteristic sets; further inputting the multiple vibration characteristic sets into a defect characteristic recognition library for similarity traversal comparison, selecting multiple defect characteristic sets that meet the dynamic comparison threshold, and performing intersection analysis on the multiple defect characteristic sets to output multiple initial defect characteristics; on the other hand, collecting the temperature distribution characteristics and acoustic wave characteristics of the target motor within a predetermined period; then performing credible probability analysis on the multiple initial defect characteristics according to the temperature distribution characteristics and acoustic wave characteristics, and outputting the initial defect characteristic with the highest probability as the detection result of the motor magnetic steel sheet defects of the target motor; that is to say, through multi-point vibration monitoring similarity comparison, multiple related potential defect characteristics can be quickly discovered, and then through joint analysis and credibility discrimination of multiple related potential defect characteristics by temperature and acoustic wave characteristics, the defect characteristic with the highest credibility is determined, thereby significantly improving the efficiency, accuracy, and reliability of motor magnetic steel sheet defect detection.

[0055] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0056] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A method for detecting defects of motor magnetic steel sheets based on vibration analysis, characterized in that, The method includes: Constructing a vibration monitoring array based on predetermined monitoring points, continuously monitoring a target motor through the vibration monitoring array, and obtaining a plurality of vibration data sets within a predetermined period; Performing vibration feature extraction on the plurality of vibration data sets to obtain a plurality of vibration feature sets; Respectively inputting the plurality of vibration feature sets into a defect feature recognition library for similarity traversal comparison, selecting a plurality of defect feature sets that meet the dynamic comparison threshold, and performing intersection analysis on the plurality of defect feature sets to output a plurality of initial defect features; Collecting the temperature distribution feature and acoustic wave feature of the target motor within a predetermined period, performing credible probability analysis on the plurality of initial defect features according to the temperature distribution feature and acoustic wave feature, and setting the initial defect feature with the highest probability as the detection result of the motor magnet steel sheet defect of the target motor; Performing vibration feature extraction on the plurality of vibration data sets to obtain a plurality of vibration feature sets, including: Performing noise point analysis and filtering and denoising processing on the plurality of vibration data sets to obtain a plurality of noise ratios and a plurality of standard vibration data sets; Configuring vibration feature indicators, where the vibration feature indicators at least include abnormal frequency components, abnormal frequency ratios, amplitude curves, and waveform curves; According to the vibration feature indicators, performing vibration feature extraction on the plurality of standard vibration data sets to obtain a plurality of vibration feature sets at a plurality of monitoring points; Respectively inputting the plurality of vibration feature sets into a defect feature recognition library for similarity traversal comparison, and selecting a plurality of defect feature sets that meet the dynamic comparison threshold, including: Constructing a defect feature recognition library based on motor attribute features, motor magnet steel sheet attribute features, and the vibration feature indicators; Calculating the mean value of the plurality of noise ratios to determine the comprehensive noise coefficient, calculating the ratio of the comprehensive noise coefficient to the standard noise coefficient, and obtaining a threshold compensation coefficient; Compensating the initial similarity comparison threshold according to the threshold compensation coefficient to obtain a dynamic comparison threshold; Respectively inputting the plurality of vibration feature sets into the defect feature recognition library for similarity traversal comparison, and selecting defect features whose similarity meets the dynamic comparison threshold, and outputting the plurality of defect feature sets, where the defect features include the defect type and defect intensity of the motor magnet steel sheet, and the defect types include cracks, detachment, warping, and wear.

2. The method for detecting defects of motor magnetic steel sheets based on vibration analysis according to claim 1, wherein, Constructing a vibration monitoring array based on predetermined monitoring points, continuously monitoring a target motor through the vibration monitoring array, and obtaining a plurality of vibration data sets within a predetermined period, including: Obtaining the predetermined monitoring points of the target motor, where the predetermined monitoring points at least include a motor stator, a motor rotor, and a motor bearing; Respectively arranging vibration sensors at the motor stator, the motor rotor, and the motor bearing to construct a vibration monitoring array; According to a predetermined monitoring frequency, continuously monitoring the target motor through the vibration monitoring array to obtain a plurality of vibration data sets at a plurality of monitoring points within a predetermined period.

3. A method for detecting defects of motor magnetic steel sheets based on vibration analysis according to claim 1, characterized in that, Constructing a defect feature recognition library based on motor attribute features, motor magnet steel sheet attribute features, and the vibration feature indicators, including: Taking the motor attribute features and the motor magnet sheet attribute features as device constraints, the vibration characteristic indexes as conditional constraints, and the motor magnet sheet defect detection based on vibration analysis as a guide, information retrieval is carried out based on big data to obtain multiple sample vibration characteristic sets, and the defect types and defect intensities of the motor magnet sheets under different sample vibration characteristic sets are labeled to obtain multiple sample defect types and multiple sample defect intensities; Based on the same defect type and defect intensity interval, clustering is performed on the multiple sample defect types and multiple sample defect intensities to determine multiple standard defect types, where each standard defect type is marked with multiple defect intensity intervals; Randomly select a first standard defect type and a first defect intensity interval of the first standard defect type, and screen the multiple sample vibration characteristic sets according to the first standard defect type and the first defect intensity interval to determine multiple first sample vibration characteristic sets; Perform high-frequency feature analysis on the multiple first sample vibration characteristic sets to determine a first standard vibration characteristic set, where the first standard vibration characteristic set includes a first standard abnormal frequency component, a first standard abnormal frequency ratio, a first standard amplitude curve, and a first standard waveform curve; Establish a first mapping association between the first standard vibration characteristic set and the first standard defect type and the first defect intensity interval, and construct a first defect recognition branch according to the first mapping association; Analyze and obtain multiple defect recognition branches in sequence, and construct the defect characteristic recognition library according to the multiple defect recognition branches; 4. A method for detecting defects of motor magnetic steel sheets based on vibration analysis according to claim 3, characterized in that, Input the multiple vibration characteristic sets into the defect characteristic recognition library respectively for similarity traversal comparison, and select defect characteristics whose similarity meets the dynamic comparison threshold, and output the multiple defect characteristic sets, including: Randomly select a first vibration characteristic set and randomly select a first defect recognition branch in the defect characteristic recognition library; Input the first vibration characteristic set into the first defect recognition branch, and perform similarity traversal comparison with the first standard vibration characteristic set. If the comparison similarity meets the dynamic comparison threshold, output the first standard defect type and the first defect intensity interval as the first defect characteristic; Sequentially perform similarity traversal comparison between the first vibration characteristic set and the standard vibration characteristic sets of other defect recognition branches in the defect characteristic recognition library to obtain a first defect characteristic set and add it to the multiple defect characteristic sets; 5. A method for detecting defects of motor magnetic steel sheets based on vibration analysis according to claim 4, characterized in that, Input the first vibration characteristic set into the first defect recognition branch, and perform similarity traversal comparison with the first standard vibration characteristic set. If the comparison similarity meets the dynamic comparison threshold, output the first standard defect type and the first defect intensity interval as the first defect characteristic, including: Obtain the first vibration characteristic set, where the first vibration characteristic set includes a first abnormal frequency component, a first abnormal frequency ratio, a first amplitude curve, and a first waveform curve; Perform similarity comparison between the first abnormal frequency component and the first standard abnormal frequency component to determine a first similarity; If the first similarity is greater than the dynamic comparison threshold, perform a similarity comparison on the first abnormal frequency ratio and the first standard abnormal frequency ratio to determine the second similarity; If the second similarity is greater than the dynamic comparison threshold, perform a similarity comparison on the first amplitude curve and the first standard amplitude curve to determine the third similarity; If the third similarity is greater than the dynamic comparison threshold, perform a similarity comparison on the first waveform curve and the first standard waveform curve to determine the fourth similarity. If the fourth similarity is greater than the dynamic comparison threshold, set the output of the first standard defect type and the first defect intensity range as the first defect feature.

6. The method for detecting defects of motor magnetic steel sheets based on vibration analysis according to claim 3, characterized in that Based on the temperature distribution feature and the acoustic wave feature, perform a credible probability analysis on the multiple initial defect features, and set the initial defect feature with the highest probability as the motor steel sheet defect detection result of the target motor, including: Taking the motor attribute feature and the motor steel sheet attribute feature as device constraints, taking the defect type and the defect intensity as feature constraints, and taking the motor steel sheet defect detection as a guide, retrieve and obtain a sample temperature distribution feature set, a sample acoustic wave feature set, and multiple sample defect feature sets, and count the proportion of sample defect features under different sample temperature distribution features and different sample acoustic wave features, set it as the trigger probability of multiple defect features, construct a sample defect feature probability distribution, and obtain a sample defect feature probability distribution set; Use the sample temperature distribution feature set, the sample acoustic wave feature set, the multiple sample defect feature sets, and the sample defect feature probability distribution set to perform supervised training on N prediction operators until convergence, obtain N defect feature trigger probability prediction units, and integrally construct a defect feature trigger probability predictor based on the N defect feature trigger probability prediction units, where the prediction operators at least include a BP neural network, a random forest, and a support vector machine, and N is an integer greater than or equal to 3; Input the temperature distribution feature, the acoustic wave feature, and the multiple initial defect features into the defect feature trigger probability predictor to perform a credible probability analysis, output a defect feature probability distribution, and select the initial defect feature with the highest probability in the defect feature probability distribution as the motor steel sheet defect detection result of the target motor.

Images