Axial magnetic field motor rotor assembly fault detection method and system

Abstract

The present invention relates to a method and system for detecting faults of an axial magnetic field motor rotor assembly, comprising: collecting vibration waveform data, temperature gradient distribution data and magnetic field intensity distribution data of a rotor assembly in an operating state to construct a dynamic monitoring data set; performing dynamic feature extraction on data in the dynamic monitoring data set to generate a dynamic operating feature set including vibration coupling features, temperature abnormal fluctuation features and magnetic field offset features, performing multi-dimensional correlation mapping processing based on the dynamic operating feature set to generate a comprehensive abnormality indicator set; determining a fault type set and a corresponding fault confidence score of a target rotor assembly based on a matching result between the comprehensive abnormality indicator set and a preset fault judgment threshold; generating a visual detection report based on the fault type set of the rotor assembly and the corresponding fault confidence score, wherein the visual detection report is used to display fault results on a terminal, thereby achieving accurate detection and intuitive presentation of rotor assembly faults.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method and system for detecting faults in an axial magnetic field motor rotor assembly. Background Art

[0002] Axial-field motors have been widely used due to their unique structural and performance advantages, such as high power density and compact axial dimensions. However, the rotor assembly of axial-field motors is prone to various faults during long-term operation due to the combined influence of mechanical stress, electromagnetic interference, thermal effects, and other factors.

[0003] Existing fault detection methods for axial magnetic field motor rotor assemblies have certain limitations. Some methods rely solely on single-dimensional monitoring data, such as vibration or temperature data. This approach struggles to fully and accurately reflect the operating status of the rotor assembly, and can easily lead to missed or misjudged faults. Furthermore, some detection methods lack effective data fusion and correlation analysis when processing multi-dimensional data, failing to fully explore the inherent connections between data from different dimensions. This results in inaccurate fault type determinations and makes it difficult to accurately predict faults in advance, posing a risk to the safe and stable operation of the equipment. Summary of the Invention

[0004] The object of the present invention is to provide a method and system for detecting faults of an axial magnetic field motor rotor assembly, aiming to detect faults of an axial magnetic field motor rotor assembly more comprehensively, accurately and efficiently.

[0005] To achieve the above objectives, a first aspect of an embodiment of the present disclosure provides a method for detecting a fault in an axial magnetic field motor rotor assembly, the method comprising:

[0006] collecting vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor when in operation, and constructing a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data;

[0007] performing dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and performing multidimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly;

[0008] Determining a fault type set and a corresponding fault confidence score of the target rotor assembly having a fault based on a matching result between the comprehensive abnormality indicator set and a preset fault determination threshold;

[0009] A visual inspection report for the target rotor assembly is generated based on the fault type set and the corresponding fault confidence score of the target rotor assembly, and the visual inspection report is used to display the fault result on the terminal.

[0010] In some possible implementations, the performing of dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features includes:

[0011] Performing frequency domain transformation processing on the vibration waveform data to generate the vibration coupling feature;

[0012] Performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and using the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature;

[0013] Performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve, and determining the magnetic field offset characteristics according to the curvature change points of the gradient curve;

[0014] The vibration coupling characteristics, temperature abnormal fluctuation characteristics and magnetic field offset characteristics are respectively subjected to sliding window normalization processing, and the normalized characteristics are associated according to the time-space dimension to generate the dynamic operation feature set containing time series and space correlation.

[0015] In some possible implementations, performing frequency domain transformation on the vibration waveform data to generate the vibration coupling feature includes:

[0016] Dividing the vibration waveform data into a plurality of time series segments of equal length, performing fast Fourier transform processing on each of the time series segments to generate corresponding vibration spectrum segments;

[0017] Extract the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of each vibration spectrum segment;

[0018] Performing sliding average processing on the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of multiple time series segments to generate a continuous vibration spectrum feature;

[0019] Setting a first target frequency band corresponding to the rotor structural resonance frequency of the rotor assembly, and calculating a first proportional coefficient of the integral value of the vibration energy in the first target frequency band to the integral value of the vibration energy in the entire frequency band;

[0020] Setting a second target frequency band corresponding to the characteristic frequency of bearing wear of the rotor assembly, and calculating a second proportional coefficient of the integral value of the vibration energy in the second target frequency band to the integral value of the vibration energy in the entire frequency band;

[0021] Setting a weight distribution ratio of the first proportional coefficient to the second proportional coefficient according to a correlation between historical fault data of the rotor structure resonance frequency and the bearing wear characteristic frequency;

[0022] The sum of the weighted first proportional coefficient and the second proportional coefficient is used as the vibration coupling feature.

[0023] In some possible implementations, performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and using the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature, includes:

[0024] Setting the length of the sliding window to a preset time period, performing a linear regression analysis on the temperature data of each temperature sensing unit within the sliding window to obtain a temperature change trend equation;

[0025] extracting a slope parameter from the temperature change trend equation as a temperature change trend slope corresponding to the temperature sensing unit;

[0026] When the temperature change trend slope exceeds a preset positive change threshold, the temperature sensing unit is marked as an abnormal temperature rising unit;

[0027] Acquiring a physical distance parameter between adjacent temperature sensing units, and calculating a temperature conduction delay compensation coefficient based on the physical distance parameter;

[0028] Performing time alignment processing on the temperature change trend slopes of adjacent temperature sensing units according to the temperature conduction delay compensation coefficient;

[0029] The absolute value of the slope difference between the aligned adjacent temperature sensing units is calculated, and the absolute value of the difference is compared with a preset difference threshold to generate the abnormal temperature fluctuation feature.

[0030] In some possible implementations, performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve includes:

[0031] Constructing a three-dimensional distribution surface of magnetic field intensity according to the spatial coordinate positions of multiple magnetic field detection devices;

[0032] Slicing the three-dimensional distribution surface along the axial direction of the rotor assembly to extract magnetic field intensity distribution data of each slice plane;

[0033] Perform polynomial fitting on the magnetic field intensity distribution data of each slice plane to generate the corresponding axial magnetic field intensity gradient curve;

[0034] The similarity coefficients between the gradient curves of adjacent slice planes are calculated, and the slice planes with similarity coefficients lower than a preset threshold are marked as magnetic field anomaly areas.

[0035] In some possible implementations, collecting vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor in an operating state, and constructing a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data, includes:

[0036] Deploying a plurality of vibration sensors at preset positions of the rotor assembly, and collecting the vibration waveform data at a first sampling period through the vibration sensors, wherein the vibration waveform data includes time-series waveforms of an axial vibration component and a radial vibration component;

[0037] A plurality of temperature sensing units are provided on the surface of the rotor assembly, and the temperature gradient distribution data is collected by the temperature sensing units at a second sampling period, wherein the temperature gradient distribution data includes temperature change rates of different regions and temperature differences between adjacent regions;

[0038] a plurality of magnetic field detection devices are provided along the circumferential direction of the rotor assembly, and the magnetic field intensity distribution data is collected by the magnetic field detection devices at a third sampling period, wherein the magnetic field intensity distribution data includes an axial magnetic field intensity distribution curve and a magnetic field asymmetry coefficient;

[0039] Performing time stamp synchronization processing on the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data, respectively, and unifying data with inconsistent sampling frequencies to the same time base using interpolation or resampling methods;

[0040] A data mapping table is established according to the spatial position relationship of the sensors, and the synchronized data are fused in a time-space alignment manner to generate the dynamic monitoring data set.

[0041] In some possible implementations, performing multi-dimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly includes:

[0042] Establishing a first correlation mapping relationship between the vibration coupling feature and the temperature abnormal fluctuation feature, and calculating a vibration-temperature coupling abnormality index based on the first correlation mapping relationship;

[0043] Establishing a second correlation mapping relationship between the temperature anomaly fluctuation characteristics and the magnetic field offset characteristics, and calculating a temperature-magnetic field interaction anomaly index based on the second correlation mapping relationship;

[0044] establishing a third association mapping relationship between the magnetic field offset feature and the vibration coupling feature, and calculating a magnetic field vibration synchronization anomaly index based on the third association mapping relationship;

[0045] The vibration-temperature coupling anomaly index, temperature-magnetic field interaction anomaly index and magnetic field-vibration synchronization anomaly index are normalized respectively, and dynamic weight coefficients are allocated according to the contribution of each type of anomaly index in the historical fault data. The normalized anomaly indexes are weighted and fused based on the dynamic weight coefficients to generate the comprehensive anomaly indicator set.

[0046] In some possible implementations, determining the fault type set and corresponding fault confidence score of the target rotor assembly based on a matching result between the comprehensive abnormality indicator set and a preset fault determination threshold value includes:

[0047] Matching the vibration-temperature coupling anomaly index with a first fault threshold interval; if the matching result is beyond the first fault threshold interval, determining that a rotor imbalance fault type exists, and calculating a first confidence score based on the excess magnitude;

[0048] Matching the temperature-magnetic field interaction anomaly index with a second fault threshold interval; if the matching result is beyond the second fault threshold interval, determining that a bearing overheating fault type exists, and calculating a second confidence score based on the excess magnitude;

[0049] Matching the magnetic field vibration synchronization anomaly index with a third fault threshold interval; if the matching result is beyond the third fault threshold interval, determining that a magnetic pole shift fault type exists, and calculating a third confidence score based on the excess magnitude;

[0050] The rotor imbalance fault type, bearing overheat fault type and magnetic pole offset fault type are combined to generate the fault type set of the target rotor assembly with the fault, and the first confidence score, the second confidence score and the third confidence score are combined to generate the fault confidence score of the target rotor assembly with the fault.

[0051] According to a second aspect of the present disclosure, a system for detecting a fault in an axial magnetic field motor rotor assembly is provided. The system comprises:

[0052] a construction module configured to collect vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor in an operating state, and construct a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data;

[0053] a first generating module configured to perform dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and perform multi-dimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly;

[0054] a determination module configured to determine a set of fault types and corresponding fault confidence scores of the target rotor assembly having a fault based on a matching result between the set of comprehensive abnormality indicators and a preset fault determination threshold;

[0055] The second generating module is configured to generate a visual detection report for the target rotor assembly according to the fault type set and the corresponding fault confidence score of the target rotor assembly, and the visual detection report is used to display the fault results on the terminal.

[0056] According to a third aspect of the present disclosure, an electronic device is provided, including:

[0057] a memory having a computer program stored thereon;

[0058] A processor is used to execute the computer program in the memory to implement the steps of any one of the methods in the first aspect.

[0059] The present invention provides a method and system for detecting faults in an axial magnetic field motor rotor assembly. Compared with the prior art, it has the following advantages:

[0060] By collecting a multi-dimensional dynamic monitoring data set, it comprehensively covers information on vibration, temperature, and magnetic field strength. Compared with single-dimensional monitoring data, it can more completely reflect the operating status of the target rotor assembly. Dynamic feature extraction and multi-dimensional correlation mapping processing are performed on the multi-dimensional data to explore the inherent connections between data of different dimensions. The generated comprehensive abnormality indicator set can more accurately reflect the abnormal conditions of the rotor assembly. Based on the matching results of the comprehensive abnormality indicator set and the preset fault judgment threshold, the fault type set and the corresponding fault confidence score are determined, which improves the accuracy and reliability of fault judgment. The visual detection report is output and transmitted to the target terminal for display, making it convenient for relevant personnel to intuitively and promptly understand the fault conditions of the rotor assembly, so that they can quickly take corresponding maintenance measures to ensure the safe and stable operation of the axial magnetic field motor and reduce the maintenance cost of the axial magnetic field motor and the losses caused by failures.

[0061] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following detailed description, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0063] Figure 1 This is a flow chart of a method for detecting faults in an axial magnetic field motor rotor assembly according to an embodiment of the specification.

[0064] Figure 2 It is a block diagram of a fault detection system for an axial magnetic field motor rotor assembly according to an embodiment of the specification.

[0065] Figure 3 The present invention is a block diagram of a device for executing a method for detecting a fault of an axial magnetic field motor rotor assembly according to an embodiment of the specification. DETAILED DESCRIPTION

[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0067] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0068] The present disclosure provides a method for detecting faults in an axial magnetic field motor rotor assembly, which is applied to a motor controller. Figure 1 This is a flow chart illustrating a method for detecting a fault in an axial magnetic field motor rotor assembly according to an embodiment. Specifically, the method includes:

[0069] In step S11, vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor are collected during operation, and a dynamic monitoring data set is constructed based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data;

[0070] Axial magnetic field motors, for example, can be used in automated guided vehicles (AGVs). They offer advantages such as compact structure, high power density, and high efficiency, providing robust and efficient power for these vehicles. The rotor assembly is the rotating part of an axial magnetic field motor and typically consists of a rotor core and rotor windings. During AGV operation, the state of the rotor assembly directly impacts the motor's performance and the vehicle's stability.

[0071] Vibration waveform data, collected by vibration sensors installed on or near the rotor assembly, reflects the time-varying vibrations of the rotor assembly during operation. This data contains rich information about the mechanical state of the rotor assembly and can be used to determine whether a mechanical fault exists.

[0072] Temperature gradient distribution data describes temperature differences at different locations within the rotor assembly. By placing temperature sensors at different locations within the rotor assembly, this data can reveal the rotor's heat conduction and localized overheating, helping to identify faults such as winding shorts and insulation damage.

[0073] Magnetic field strength distribution data represents the spatial distribution of magnetic field strength around the rotor assembly. Measured using a magnetic field sensor, it reflects the motor's electromagnetic performance and helps identify issues such as magnetic field distortion and pole demagnetization.

[0074] In the disclosed embodiments, during the operation of an AGV, the rotor assembly of an axial magnetic field motor generates vibrations, temperature fluctuations, and magnetic field variations. By strategically placing vibration sensors, temperature sensors, and magnetic field sensors around the rotor assembly, real-time data on vibration waveforms, temperature gradient distribution, and magnetic field intensity distribution are collected. This data is organized and stored chronologically and spatially to construct a dynamic monitoring data set, enabling subsequent analysis and diagnosis of the rotor assembly's operating status.

[0075] For example, on the axial magnetic field motor of an automated guided vehicle (AGV), three vibration sensors are installed at different locations on the rotor assembly (such as the top, middle, and bottom). Two temperature sensors measure the temperature of the rotor windings and core, respectively, and a magnetic field sensor measures the magnetic field strength surrounding the rotor. After the AGV is started and operates for a period of time, the vibration sensors collect vibration waveform data at different locations, the temperature sensors record temperature changes in the windings and core, and the magnetic field sensor captures magnetic field strength distribution data. These data, when combined, form a dynamic monitoring data set for the rotor assembly during that period.

[0076] In step S12, dynamic feature extraction processing is performed on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and multi-dimensional correlation mapping processing is performed based on the dynamic operation feature set to generate a comprehensive abnormality indicator set of the rotor assembly;

[0077] Vibration coupling characteristics are characterized by the mutual coupling and influence between different frequency components in a vibration signal. In the axial magnetic field motor of an automated guided vehicle (AGV), vibration coupling characteristics can reflect the interaction between multiple mechanical components in the rotor assembly (such as bearings and rotor core), helping to identify complex mechanical faults.

[0078] Abnormal temperature fluctuations are unusual changes in temperature data, such as sudden temperature increases or decreases, or large fluctuations. During AGV operation, these fluctuations may indicate problems such as localized overheating, poor heat dissipation, or electrical failures in the rotor assembly.

[0079] A magnetic field excursion signature is a deviation in magnetic field intensity distribution data compared to the normal state. In axial field motors, a magnetic field excursion signature can indicate faults such as abnormal magnetic pole position, magnetic path blockage, or magnetic pole demagnetization.

[0080] In the disclosed embodiments, dynamic feature extraction is performed on the data in the dynamic monitoring data set. Signal processing techniques (such as Fourier transform and wavelet analysis) are used to extract vibration coupling features from the vibration waveform data, identify abnormal temperature fluctuation features from the temperature gradient distribution data, and detect magnetic field offset features from the magnetic field intensity distribution data. Based on the extracted dynamic operation feature set, data mining and machine learning algorithms (such as association rule mining and neural networks) are then used to perform multidimensional association mapping, analyze the relationships between different features, and generate a comprehensive set of abnormality indicators that can comprehensively reflect the operating status of the rotor assembly.

[0081] For example, the collected vibration waveform data is converted from the time domain to the frequency domain through Fourier transform, and the amplitude and phase relationship of different frequency components are analyzed to extract vibration coupling characteristics, such as the harmonic relationship or coupling frequency between certain frequency components.

[0082] For temperature gradient distribution data, a temperature change threshold is set. When the temperature change exceeds this threshold, it is determined to be an abnormal temperature fluctuation feature. For magnetic field intensity distribution data, the deviation of magnetic field intensity from the normal value at different locations is calculated to obtain the magnetic field offset feature. Then, an association rule mining algorithm is used to analyze the correlation between vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features. For example, it is found that when harmonics of a specific frequency appear in the vibration coupling feature and the abnormal temperature fluctuation feature manifests as a sudden temperature increase, the magnetic field offset feature will also increase accordingly. Based on these correlations, a set of comprehensive anomaly indicators is generated, such as the comprehensive anomaly index and the fault risk level.

[0083] In step S13, based on the matching result between the comprehensive abnormality indicator set and the preset fault judgment threshold, the fault type set and the corresponding fault confidence score of the target rotor assembly with the fault are determined;

[0084] In the disclosed embodiment, the generated set of comprehensive anomaly indicators is compared and matched with preset fault determination thresholds. Based on the matching results, the presence of a rotor assembly fault and the possible fault type are determined. For each possible fault type, a corresponding fault confidence score is assigned based on its correlation with the comprehensive anomaly indicators and its severity. The fault confidence score can be determined through methods such as expert experience, historical data statistics, or machine learning models.

[0085] For example, assume that in the preset fault judgment threshold, a fault is considered to exist when the comprehensive anomaly index is greater than 80. In the generated comprehensive anomaly indicator set, the comprehensive anomaly index is 85, which exceeds the fault judgment threshold, and therefore it is determined that the rotor assembly has a fault. Further analysis of the association between the comprehensive anomaly indicator and different fault types shows that when the comprehensive anomaly index is between 80-90, the possibility of a bearing fault is greater, and the possibility of a winding short circuit is second. According to historical data statistics, in this case, the confidence score for the bearing fault is 0.8, and the confidence score for the winding short circuit is 0.3. Therefore, the fault type set of the rotor assembly is determined to be {bearing fault, winding short circuit}, and the corresponding fault confidence score is {0.8, 0.3}.

[0086] In step S14, a visual inspection report for the target rotor assembly is generated according to the fault type set and the corresponding fault confidence score of the target rotor assembly. The visual inspection report is used to display the fault result on the terminal.

[0087] The visual inspection report presents the rotor assembly fault types and corresponding fault confidence scores in an intuitive and visual manner. The report may include charts, graphs, and other elements to facilitate the display and analysis of fault results on the terminal.

[0088] In the disclosed embodiment, a visual inspection report is generated using visualization techniques (such as charts, graphs, and reports) based on the determined set of rotor assembly fault types and their corresponding fault confidence scores. The report can include a bar chart displaying the confidence scores for different fault types, a line chart showing the changing trends of comprehensive abnormality indicators over time, and detailed fault descriptions and recommended measures. The generated visual inspection report is then transmitted to a terminal device, allowing users to conveniently display and analyze fault results.

[0089] For example, using Python's Matplotlib library to generate a visual inspection report includes a bar chart with the fault type (bearing fault, winding short circuit) plotted on the horizontal axis and the fault confidence score on the vertical axis. The bar chart visually shows that the confidence score for the bearing fault is 0.8, while the confidence score for the winding short circuit is 0.3.

[0090] The report also includes a line graph showing the changing trend of the comprehensive abnormality index over time, allowing users to understand the changes in the rotor assembly's operating status. The report also includes a detailed description of the fault, including possible causes of bearing failure (bearing wear, insufficient lubrication, etc.) and recommended measures (bearing replacement, increased lubrication, etc.). This report is transmitted to the AGV's monitoring terminal, where operators can clearly view the fault information and take appropriate measures in a timely manner.

[0091] The above technical solution comprehensively covers information on vibration, temperature, and magnetic field strength by collecting a multi-dimensional dynamic monitoring data set. Compared with single-dimensional monitoring data, it can more completely reflect the operating status of the target rotor assembly. Dynamic feature extraction and multi-dimensional correlation mapping processing are performed on multi-dimensional data to explore the inherent connections between data of different dimensions. The generated comprehensive abnormality indicator set can more accurately reflect the abnormal conditions of the rotor assembly. The fault type set and the corresponding fault confidence score are determined based on the matching results of the comprehensive abnormality indicator set and the preset fault judgment threshold, thereby improving the accuracy and reliability of fault judgment. The visual detection report is output and transmitted to the target terminal for display, so that relevant personnel can intuitively and promptly grasp the fault conditions of the rotor assembly, so that they can quickly take corresponding maintenance measures to ensure the safe and stable operation of the axial magnetic field motor and reduce the maintenance cost of the axial magnetic field motor and the losses caused by failures.

[0092] In a possible implementation, in step S12, the data in the dynamic monitoring data set are subjected to dynamic feature extraction processing to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, including:

[0093] In step S121, frequency domain transformation processing is performed on the vibration waveform data to generate the vibration coupling feature;

[0094] In the disclosed embodiments, the vibration waveform data is a time-varying time-domain signal, making it difficult to directly extract features reflecting the relationships between different frequency components. A frequency domain transformation (such as a Fourier transform) is used to convert the time-domain vibration signal to the frequency domain, obtaining the signal's amplitude and phase information at different frequencies. In the frequency domain, coupling relationships between different frequency components, such as harmonic coupling and sideband coupling, can be analyzed. These coupling relationships reflect the mechanical interactions between various components in the rotor assembly (such as bearings and rotor core), thereby generating vibration coupling features.

[0095] For example, a vibration sensor is installed on the axial magnetic field motor of an automated guided vehicle (AGV) to collect vibration waveform data of the rotor assembly during operation. Fast Fourier transform (FFT) is used to convert the time-domain vibration signal into a frequency-domain signal. Analysis of the frequency-domain signal reveals additional amplitude peaks at integer multiples of the fundamental frequency (e.g., the motor's rotational frequency). These peaks exhibit a phase relationship, indicating harmonic coupling. Furthermore, sidebands related to the fundamental frequency are observed. The frequency spacing of these sidebands correlates with the characteristic frequency of bearing faults, further demonstrating the mechanical coupling between the bearing and the rotor. These characteristics, such as harmonic coupling and sidebands, are considered vibration coupling signatures.

[0096] In step S122, a sliding window statistical analysis is performed on the temperature gradient distribution data to calculate the temperature change trend slope corresponding to each temperature sensing unit, and the difference between the temperature change trend slopes of adjacent temperature sensing units is used as the temperature abnormal fluctuation feature;

[0097] Among them, sliding window statistical analysis can set a fixed-size window to slide on the data sequence and perform statistical analysis on the data within the window to capture the local change trend of the data.

[0098] In the disclosed embodiment, the temperature gradient distribution data reflects the temperature changes at different positions of the motor rotor assembly. Through sliding window statistical analysis, the temperature data is linearly fitted in each sliding window, and the slope of the temperature change trend over time in the window, that is, the temperature change rate, is calculated. The temperature change trend slopes of adjacent temperature sensing units should have a certain degree of similarity. If the difference between adjacent slopes is too large, it means that the temperature change in this area has abnormally fluctuated, which may be caused by local overheating, poor heat dissipation, or electrical failure. This difference value is used as a characteristic of abnormal temperature fluctuation.

[0099] For example, multiple temperature sensors are evenly distributed on the rotor of the axial magnetic field motor of an automated guided vehicle (AGV) to collect temperature gradient distribution data. The sliding window size is set to 10 data points, and the window slides one data point at a time. For each temperature sensing unit, a least squares linear fit is performed within each sliding window to obtain the temperature change trend slope. For example, at a certain moment, the temperature change trend slope within the current sliding window of temperature sensing unit A is 0.5°C / s, while the slope of the adjacent temperature sensing unit B is -0.2°C / s, with a difference of 0.7°C / s between the two. If this difference exceeds a preset threshold (e.g., 0.3°C / s), it is considered that an abnormal temperature fluctuation exists in that area, and this difference is included as part of the temperature fluctuation feature.

[0100] In step S123, spatial distribution fitting is performed on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve, and the magnetic field offset feature is determined according to the curvature change point of the gradient curve;

[0101] Spatial distribution fitting utilizes mathematical methods to fit magnetic field intensity distribution data to obtain a continuous spatial distribution model of magnetic field intensity, facilitating analysis of magnetic field characteristics. The axial magnetic field intensity gradient curve, generated by spatial distribution fitting of magnetic field intensity distribution data, describes the axial variation of magnetic field intensity. Curvature change points are points in the axial magnetic field intensity gradient curve where the curvature changes significantly. These points may correspond to anomalies in the magnetic field distribution and are used to identify magnetic field offset characteristics.

[0102] In the disclosed embodiments, the magnetic field intensity distribution data is represented by discrete spatial data points. These discrete data points are fitted into a continuous magnetic field intensity distribution model using spatial distribution fitting methods (such as polynomial fitting and spline interpolation). The fitted magnetic field intensity distribution is then differentiated along the axial direction to produce an axial magnetic field intensity gradient curve, which describes the axial variation of the magnetic field intensity. Under normal circumstances, the magnetic field intensity gradient curve should be smooth. However, when a motor experiences a magnetic circuit fault (such as pole demagnetization or magnetic circuit blockage), the magnetic field intensity distribution may change abnormally, resulting in curvature change points in the gradient curve. By detecting these curvature change points, the magnetic field offset characteristics can be determined.

[0103] For example, magnetic field sensors are placed around the axial magnetic field motor of an automated guided vehicle (AGV) to collect magnetic field intensity distribution data. A quadratic polynomial fitting method is used to fit the magnetic field intensity distribution data, resulting in a continuous spatial distribution function for the magnetic field intensity. This function is then differentiated along the axial direction to generate an axial magnetic field intensity gradient curve. Analysis of the gradient curve reveals a significant change in curvature at a certain axial position, with an inflection point appearing in the previously smooth curve. The axial position corresponding to this inflection point may indicate a shift in the magnetic field distribution at that location, possibly due to weakened magnetism at a particular magnetic pole or the presence of foreign matter in the magnetic circuit. The location and associated characteristics of this curvature change point are used as magnetic field offset features.

[0104] In step S124, the vibration coupling features, temperature abnormal fluctuation features and magnetic field offset features are respectively subjected to sliding window normalization processing, and the normalized features are associated according to the time-space dimension to generate the dynamic operation feature set containing time sequence and space correlation.

[0105] Among them, sliding window normalization is to normalize the extracted feature data so that they have the same dimension and distribution characteristics, which is convenient for subsequent comparison and analysis. Sliding window normalization is to normalize the data within the sliding window.

[0106] In the embodiment of the present disclosure, since the vibration coupling characteristics, temperature abnormal fluctuation characteristics, and magnetic field offset characteristics may have different dimensions and distribution characteristics, direct comparison and analysis may be affected by dimensional differences. Therefore, a sliding window normalization process is used to normalize the data of each feature within the sliding window (such as Z-score normalization) so that they have the same mean and standard deviation. Then, these standardized features are associated according to the time and space dimensions. The time dimension association can reflect the changing trend of the feature over time, and the space dimension association can reflect the distribution of the feature at different locations, thereby generating a dynamic operation feature set that contains time series and spatial correlations, providing more comprehensive information for subsequent fault diagnosis.

[0107] For example, for the extracted vibration coupling features, temperature anomaly fluctuation features, and magnetic field offset features, the sliding window size was set to 20 data points, and the window slid by 5 data points at a time. Within each sliding window, the vibration coupling features were normalized using the Z-score formula z = σx - μ, where x is the current eigenvalue, μ is the mean of the eigenvalues ​​within the window, and σ is the standard deviation of the eigenvalues ​​within the window. Similarly, the temperature anomaly fluctuation features and magnetic field offset features were normalized in the same way.

[0108] In the time dimension, the standardized features are arranged in chronological order to analyze the changing trends of the features over time. For example, observing the fluctuations of the vibration coupling features over a period of time to determine whether there is a gradually increasing trend may indicate the gradual development of a fault. In the spatial dimension, the standardized features of different locations (such as different parts of the rotor) are correlated. For example, comparing the abnormal temperature fluctuation characteristics of adjacent locations. If the characteristics of adjacent locations differ greatly, it may indicate the presence of a localized fault in that area. The features correlated in the time and space dimensions are integrated together to generate a dynamic operation feature set, such as a time series matrix containing multiple features, where each element corresponds to the feature value of a specific time and space location.

[0109] In a possible implementation, in step S121, performing frequency domain transformation processing on the vibration waveform data to generate the vibration coupling feature includes:

[0110] Dividing the vibration waveform data into a plurality of time series segments of equal length, performing fast Fourier transform processing on each of the time series segments to generate corresponding vibration spectrum segments;

[0111] In the disclosed embodiment, the collected vibration waveform data V is divided into multiple equal-length time segments Vi, where i = 1, 2, ..., n, according to a fixed length L. A fast Fourier transform is performed on each time segment Vi, converting the vibration signal in the time domain to the frequency domain to obtain the corresponding vibration spectrum segment Si. This transform can be used to analyze the vibration components at different frequencies.

[0112] Extract the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of each vibration spectrum segment;

[0113] In the disclosed embodiment, the fundamental frequency component amplitude Af, the harmonic component amplitude Ahj (where j represents the order of the harmonic), and the high-frequency noise energy value En are determined from each vibration spectrum segment Si. The fundamental frequency component amplitude reflects the strength of the primary frequency component of the vibration, the harmonic component amplitude reflects the vibration intensity of different harmonic frequencies, and the high-frequency noise energy value characterizes the impact of high-frequency noise on the vibration. For example, in Si, the amplitude Af corresponding to the fundamental frequency f and the amplitude Ahj corresponding to each harmonic frequency fj are determined, and the energy value of the high-frequency region is calculated as En.

[0114] Performing sliding average processing on the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of multiple time series segments to generate a continuous vibration spectrum feature;

[0115] In the disclosed embodiment, a sliding average is performed on the fundamental frequency component amplitude Af, harmonic component amplitude Ahj, and high-frequency noise energy value En extracted from multiple time series segments. The sliding window length is set to m, and for the fundamental frequency component amplitude, a sliding average value Afa = (Afk + Afk-1 + ... + Afk-m+1) / m is calculated, where k = m, m+1, ..., n. Similarly, sliding average values ​​of the harmonic component amplitudes and high-frequency noise energy values ​​are calculated to generate a continuous vibration spectrum feature Sc, which more smoothly displays the changes in the vibration spectrum over time.

[0116] Setting a first target frequency band corresponding to the rotor structural resonance frequency of the rotor assembly, and calculating a first proportional coefficient of the integral value of the vibration energy in the first target frequency band to the integral value of the vibration energy in the entire frequency band;

[0117] In this disclosed embodiment, a first target frequency band, Fr1, corresponding to the rotor's structural resonant frequency is determined based on the rotor's structural parameters and material properties. The integral value of the vibration energy within this frequency band, Er1, is calculated, while the integral value of the vibration energy across the entire frequency band, Ea, is calculated. A first proportional coefficient, Pr1 = Er1 / Ea, reflects the proportion of the vibration energy within the first target frequency band to the total vibration energy.

[0118] Setting a second target frequency band corresponding to the characteristic frequency of bearing wear of the rotor assembly, and calculating a second proportional coefficient of the integral value of the vibration energy in the second target frequency band to the integral value of the vibration energy in the entire frequency band;

[0119] In this disclosed embodiment, a second target frequency band Fr2, corresponding to the characteristic frequency of bearing wear, is determined based on the bearing model, material, and operating conditions. The integral value Er2 of the vibration energy within this frequency band is calculated, and a second proportional coefficient Pr2 = Er2 / Ea is used to represent the proportion of the vibration energy within the second target frequency band to the total vibration energy.

[0120] Setting a weight distribution ratio of the first proportional coefficient to the second proportional coefficient according to a correlation between historical fault data of the rotor structure resonance frequency and the bearing wear characteristic frequency;

[0121] In this disclosed embodiment, a large amount of historical rotor failure data was analyzed to determine the correlation between the rotor structural resonant frequency, the bearing wear characteristic frequency, and various types of failures. For example, it was found that the probability of a failure being caused by the rotor structural resonant frequency is w1, the probability of a failure being caused by the bearing wear characteristic frequency is w2, and that w1 + w2 = 1. Based on this, the weight of the first proportional coefficient Pr1 is set to w1, and the weight of the second proportional coefficient Pr2 is set to w2.

[0122] The sum of the weighted first proportional coefficient and the second proportional coefficient is used as the vibration coupling feature.

[0123] In the embodiment of the present disclosure, the vibration coupling characteristic Vc=w1×Pr1+w2×Pr2, which comprehensively reflects the vibration coupling related to the rotor structure resonance and bearing wear.

[0124] In one possible implementation, in step S122, performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and using the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature, includes:

[0125] Setting the length of the sliding window to a preset time period, performing a linear regression analysis on the temperature data of each temperature sensing unit within the sliding window to obtain a temperature change trend equation;

[0126] In the disclosed embodiment, a sliding window length of Tw is set. For each temperature sensing unit's collected temperature data, Tt, a linear regression analysis is performed within the sliding window of length Tw. Using methods such as the least squares method, a linear equation T = at + b is found, where a and b are coefficients determined by the regression analysis. This linear equation represents the temperature variation trend of the temperature sensing unit within the sliding window.

[0127] extracting a slope parameter from the temperature change trend equation as a temperature change trend slope corresponding to the temperature sensing unit;

[0128] In the embodiment of the present disclosure, the slope parameter a is extracted from the temperature change trend equation T=at+b. The slope a is the temperature change trend slope St corresponding to the temperature sensing unit, which reflects the rate of change of the temperature in the area where the temperature sensing unit is located over time.

[0129] When the temperature change trend slope exceeds a preset positive change threshold, the temperature sensing unit is marked as an abnormal temperature rising unit;

[0130] In the embodiment of the present disclosure, a preset positive change threshold is set as Th. If the temperature change trend slope St corresponding to a temperature sensing unit is greater than Th, the temperature sensing unit is marked as an abnormal temperature rising unit, indicating that the temperature in the area is rising too fast and there may be an abnormality.

[0131] Acquiring a physical distance parameter between adjacent temperature sensing units, and calculating a temperature conduction delay compensation coefficient based on the physical distance parameter;

[0132] In the disclosed embodiment, the physical distance d between adjacent temperature sensing units is measured. Based on thermal conduction theory and material properties, a temperature conduction delay compensation coefficient Ct is calculated. For example, Ct may be related to factors such as distance d and material thermal conductivity. This coefficient is calculated using a related formula to account for the delay effect of temperature conduction between adjacent areas.

[0133] Performing time alignment processing on the temperature change trend slopes of adjacent temperature sensing units according to the temperature conduction delay compensation coefficient;

[0134] In the disclosed embodiment, the temperature conduction delay compensation coefficient Ct is used to time-align the temperature change trend slopes St1 and St2 of adjacent temperature sensing units. For example, by adjusting the time coordinates or data processing methods, the slope changes caused by the temperature conduction delay are synchronized in time, ensuring the accuracy of the subsequent difference calculation.

[0135] The absolute value of the slope difference between the aligned adjacent temperature sensing units is calculated, and the absolute value of the difference is compared with a preset difference threshold to generate the abnormal temperature fluctuation feature.

[0136] In this disclosed embodiment, the absolute value of the slope difference between adjacent aligned temperature sensing units, D = |St1 - St2|, is calculated. This difference, D, is compared with a preset difference threshold, Td. If D > Td, it indicates significant temperature variation trends in adjacent areas, indicating the presence of abnormal temperature fluctuations. This generates the abnormal temperature fluctuation signature, Ta.

[0137] In a possible implementation, in step S123, performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve includes:

[0138] Constructing a three-dimensional distribution surface of magnetic field intensity according to the spatial coordinate positions of multiple magnetic field detection devices;

[0139] In the embodiment of the present disclosure, a three-dimensional distribution surface S (x, y, z) of the magnetic field intensity is constructed using a surface fitting algorithm based on the spatial coordinate positions (x, y, z) of multiple magnetic field detection devices in the circumferential and axial directions of the rotor assembly, and the magnetic field strength values ​​B they collect. This surface comprehensively reflects the distribution of the magnetic field in space.

[0140] Slicing the three-dimensional distribution surface along the axial direction of the rotor assembly to extract magnetic field intensity distribution data of each slice plane;

[0141] In the disclosed embodiment, the three-dimensional magnetic field intensity distribution surface S(x, y, z) is sliced ​​at regular intervals along the axial direction of the rotor assembly. For example, slices are taken at intervals of Δz to produce a series of slice planes. From each slice plane, magnetic field intensity distribution data B(x, y) is extracted. This data reflects the distribution of magnetic field intensity on the plane at different axial positions.

[0142] Perform polynomial fitting on the magnetic field intensity distribution data of each slice plane to generate the corresponding axial magnetic field intensity gradient curve;

[0143] In the embodiments of the present disclosure, for the magnetic field intensity distribution data B(x, y) of each slice plane, a polynomial fitting method is adopted, such as quadratic polynomial or cubic polynomial fitting. By fitting, a curve equation B = ax² + bx + c (taking the quadratic polynomial as an example) is obtained. This curve is the corresponding axial magnetic field intensity gradient curve Bg(x), which shows the gradient change of the magnetic field intensity along a certain direction on this slice plane.

[0144] Calculate the similarity coefficient between the gradient curves of adjacent slice planes, and mark the slice plane with a similarity coefficient lower than the preset threshold as the magnetic field abnormal area.

[0145] In the embodiments of the present disclosure, through a specific similarity calculation method, such as calculating the Euclidean distance or correlation coefficient between two curves, etc., calculate the similarity coefficient Sc between the gradient curves Bg1(x) and Bg2(x) of adjacent slice planes. Set the preset threshold as Ts. If Sc < Ts, then mark this slice plane as the magnetic field abnormal area, indicating that the magnetic field distribution in this area is quite different from that of the adjacent area, and there may be abnormal situations such as magnetic field offset. According to the curvature change points of the gradient curve, for example, by calculating the curvature through derivation, find the points where the curvature changes significantly, and determine the magnetic field offset feature Mo.

[0146] In a possible implementation manner, in step S11, collect the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor during the running state, and construct a dynamic monitoring data set according to the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data, including:

[0147] In step S111, deploy a plurality of vibration sensors at preset positions of the rotor assembly, and collect the vibration waveform data through the vibration sensors with a first sampling period. The vibration waveform data includes the time series waveforms of the axial vibration component and the radial vibration component.

[0148] In the embodiments of the present disclosure, on the rotor assembly, according to its mechanical structure characteristics and past high-fault areas, select specific preset positions to arrange vibration sensors. For example, select several positions at equal intervals at both ends and in the middle of the rotor shaft, and install a plurality of high-precision vibration sensors. Collect the vibration waveform data with a first sampling period Ta, and Ta is determined according to the running speed and vibration frequency characteristics of the motor. When the vibration sensor works, it records the waveform data of the axial vibration component Ax and the radial vibration component Ry changing with time t in real time. For example, at a certain moment t1, the vibration sensor records the axial vibration component Ax(t1) and the radial vibration component Ry(t1), and continuous collection forms continuous time series waveform data.

[0149] In step S112, a plurality of temperature sensing units are provided on the surface of the rotor assembly, and the temperature gradient distribution data is collected by the temperature sensing units at a second sampling period, wherein the temperature gradient distribution data includes the temperature change rate of different regions and the temperature difference between adjacent regions;

[0150] In the disclosed embodiment, a plurality of temperature sensing units are arranged on the surface of the rotor assembly according to a certain distribution pattern. For example, they are arranged in a uniform matrix to ensure coverage of key areas. Temperature data is collected through these temperature sensing units with a second sampling period Tb. Tb needs to comprehensively consider the temperature change rate and the requirements for time resolution of subsequent data analysis. The temperature sensing unit collects the temperature value T of each area, from which the temperature change rate Tr=(T(t+Δt)-T(t)) / Δt of different areas can be calculated, where Δt is the set time interval. At the same time, for adjacent temperature sensing units, the temperature difference Td=|T1-T2| of adjacent areas can be calculated, where T1 and T2 are the temperature values ​​collected by two adjacent temperature sensing units, respectively. These data together constitute the temperature gradient distribution data.

[0151] In step S113, a plurality of magnetic field detection devices are provided along the circumferential direction of the rotor assembly, and the magnetic field intensity distribution data is collected by the magnetic field detection devices at a third sampling period, wherein the magnetic field intensity distribution data includes an axial magnetic field intensity distribution curve and a magnetic field asymmetry coefficient;

[0152] In the embodiment of the present disclosure, a plurality of magnetic field detection devices are arranged at equal angular intervals along the circumferential direction of the rotor assembly. These devices are used to detect the magnetic field information generated when the rotor is in operation. The magnetic field intensity distribution data is collected at a third sampling period Tc, and Tc is determined based on the fluctuation characteristics of the magnetic field and the measurement accuracy requirements. The data collected by the magnetic field detection device can be processed to generate an axial magnetic field intensity distribution curve Az, which reflects the change in magnetic field intensity along the axial position x. At the same time, the magnetic field asymmetry coefficient Ac is calculated by a specific algorithm, for example, by comparing the difference in magnetic field intensity at different angular positions, etc., to obtain this coefficient, so as to measure the degree of symmetry of the magnetic field in the circumferential direction.

[0153] In step S114, the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data are time-stamped and synchronized respectively, and data with inconsistent sampling frequencies are unified to the same time base using interpolation or resampling methods;

[0154] In the disclosed embodiments, since the sampling frequencies of the vibration sensor, temperature sensing unit, and magnetic field detection device may differ, it is necessary to timestamp each of the three types of data to indicate the precise time of data acquisition. If the sampling frequency Fa of the vibration waveform data is higher than the sampling frequency Fb of the temperature gradient distribution data, an interpolation method can be employed. For example, linear interpolation can be used to insert appropriate data points within the time interval of the temperature gradient distribution data to align the time interval of the temperature data with that of the vibration waveform data, thereby unifying the three types of data to the same time reference.

[0155] In step S115 , a data mapping table is established according to the spatial position relationship of the sensors, and the synchronized data are fused in a time-space alignment manner to generate the dynamic monitoring data set.

[0156] In the disclosed embodiment, a data mapping table is constructed based on the spatial positions of the vibration sensor, temperature sensing unit, and magnetic field detection device on the rotor assembly. The specific position coordinates of each sensor are recorded. The data after timestamp synchronization processing is fused according to the time-space alignment method. That is, at the same time t, the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the corresponding positions are integrated together to form a multi-dimensional dynamic monitoring data set M. The dynamic monitoring data set can fully reflect the operating status of the rotor assembly at time t.

[0157] In a possible implementation, in step S12, performing multi-dimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set of the rotor assembly includes:

[0158] In step S121, a first correlation mapping relationship between the vibration coupling feature and the temperature abnormal fluctuation feature is established, and a vibration-temperature coupling abnormality index is calculated based on the first correlation mapping relationship;

[0159] In the embodiment of the present disclosure, the potential connection between the vibration coupling feature Vc and the abnormal temperature fluctuation feature Ta is analyzed, and a first correlation mapping relationship is established through statistical analysis of historical data or theoretical deduction. For example, it is found that changes in the vibration coupling feature will affect the abnormal temperature fluctuation to a certain extent, and there may be a linear or nonlinear relationship. Assuming that there is a linear relationship, the vibration-temperature coupling anomaly index VTa=k1×Vc+k2×Ta, where k1 and k2 are coefficients determined by methods such as regression analysis of a large amount of historical fault data, reflecting the influence weight of the vibration coupling feature and the abnormal temperature fluctuation feature on the index.

[0160] In step S122, a second correlation mapping relationship between the temperature anomaly fluctuation characteristics and the magnetic field offset characteristics is established, and a temperature-magnetic field interaction anomaly index is calculated based on the second correlation mapping relationship;

[0161] In the disclosed embodiment, the relationship between the abnormal temperature fluctuation characteristic Ta and the magnetic field offset characteristic Mo is studied, and a second correlation mapping relationship is established through experimental data or theoretical models. For example, it may be found that there is a certain causal relationship between the abnormal temperature fluctuation and the magnetic field offset. Assuming that there is a nonlinear relationship, the temperature-magnetic field interaction anomaly index TMo=f(Ta,Mo) is obtained by function fitting. The form of the function f is determined by analyzing historical data. For example, f may be a polynomial function containing Ta and Mo. The index reflects the abnormal interaction between temperature and magnetic field.

[0162] In step S123, a third correlation mapping relationship between the magnetic field offset feature and the vibration coupling feature is established, and a magnetic field vibration synchronization anomaly index is calculated based on the third correlation mapping relationship;

[0163] In the embodiment of the present disclosure, the intrinsic connection between the magnetic field offset feature Mo and the vibration coupling feature Vc is explored, and a third correlation mapping relationship is established with the help of data analysis tools and theoretical knowledge. For example, it is found that the magnetic field offset will cause the vibration characteristics to change, and vice versa. Assuming the existence of a complex nonlinear relationship, a relationship is established through methods such as a neural network model. The magnetic field vibration synchronization anomaly index MVo=g(Mo,Vc), the function g is obtained by neural network training, the input is the magnetic field offset feature Mo and the vibration coupling feature Vc, and the output is the magnetic field vibration synchronization anomaly index, which reflects the synchronization anomaly between the magnetic field and the vibration.

[0164] In step S124, the vibration-temperature coupling anomaly index, the temperature-magnetic field interaction anomaly index and the magnetic field-vibration synchronization anomaly index are normalized respectively, and dynamic weight coefficients are allocated according to the contribution of each type of anomaly index in the historical fault data. The normalized anomaly indexes are weighted and fused based on the dynamic weight coefficients to generate the comprehensive anomaly indicator set.

[0165] In the embodiments of the present disclosure, the vibration-temperature coupling anomaly index VTa, the temperature-magnetic field interaction anomaly index TMo, and the magnetic field-vibration synchronization anomaly index MVo are respectively normalized, and their values are mapped to the same value range, such as [0, 1]. By analyzing a large amount of historical fault data, the contribution degrees of various anomaly indexes in different fault types are determined. Suppose the contribution degree of the vibration-temperature coupling anomaly index in a certain type of fault is wVTa, the contribution degree of the temperature-magnetic field interaction anomaly index is wTMo, and the contribution degree of the magnetic field-vibration synchronization anomaly index is wMVo, and wVTa + wTMo + wMVo = 1. Dynamic weight coefficients are assigned according to these contribution degrees. The normalized anomaly indexes are weighted and fused, and the comprehensive anomaly index set I = wVTa × VTa_norm + wTMo × TMo_norm + wMVo × MVo_norm, where VTa_norm, TMo_norm, and MVo_norm are the normalized vibration-temperature coupling anomaly index, temperature-magnetic field interaction anomaly index, and magnetic field-vibration synchronization anomaly index respectively. This set comprehensively reflects the comprehensive anomaly situation of the rotor assembly.

[0166] In a possible implementation manner, in step S13, the determining of the fault type set and the corresponding fault confidence score of the target rotor assembly according to the matching result between the comprehensive anomaly index set and a preset fault determination threshold includes:

[0167] In step S131, the vibration-temperature coupling anomaly index is matched with a first fault threshold interval. If the matching result is that it exceeds the first fault threshold interval, it is determined that there is a rotor imbalance fault type, and a first confidence score is calculated based on the exceeding amplitude;

[0168] In the embodiments of the present disclosure, the vibration-temperature coupling anomaly index VTa is compared with a first fault threshold interval [L1, U1]. If VTa > U1 or VTa < L1, it is determined that there is a rotor imbalance fault type. A first confidence score C1 is calculated based on the exceeding amplitude. For example, if the degree to which VTa exceeds the upper limit U1 is ΔVTa = VTa - U1, C1 is calculated through a preset functional relationship f1(ΔVTa). The function f1 may be a monotonically increasing function. The greater the exceeding amplitude, the higher the first confidence score C1, indicating a greater possibility of rotor imbalance fault.

[0169] In step S132, the temperature-magnetic field interaction anomaly index is matched with a second fault threshold interval. If the matching result is that it exceeds the second fault threshold interval, it is determined that there is a bearing overheat fault type, and a second confidence score is calculated based on the exceeding amplitude;

[0170] In this disclosed embodiment, the temperature-magnetic field interaction anomaly index TMo is compared to the second fault threshold interval [L2, U2]. L2 and U2 are defined based on extensive historical axial magnetic field motor operating data and bearing overheating failure cases. This reasonable range was determined by analyzing the temperature-magnetic field interaction anomaly index of numerous similar motor models under normal operation and bearing overheating failure conditions.

[0171] Furthermore, when TMo is greater than U2 or less than L2, a bearing overheat fault type is determined. Next, a second confidence score, C2, is calculated based on the magnitude of the excess. Assuming the magnitude of the excess is represented by ΔTMo, if TMo is greater than U2, then ΔTMo is equal to TMo minus U2; if TMo is less than L2, then ΔTMo is equal to L2 minus TMo.

[0172] Furthermore, C2 can be calculated using a specific function relationship, f2, which is determined based on the relationship between the magnitude of the temperature-magnetic field interaction anomaly index exceeding the range in historical data and the actual probability of a bearing overheating failure. For example, this function may be derived from multiple data fits, and its form may be similar to a polynomial function or other function determined based on the data characteristics. The goal is to ensure that the greater the magnitude of the excess, the higher the C2 value, thereby reflecting the likelihood of a bearing overheating failure.

[0173] In step S133, the magnetic field vibration synchronization anomaly index is matched with a third fault threshold interval. If the matching result exceeds the third fault threshold interval, it is determined that a magnetic pole shift fault type exists, and a third confidence score is calculated based on the excess magnitude.

[0174] In this disclosed embodiment, the magnetic field vibration synchronization anomaly index MVo is compared with the third fault threshold interval [L3, U3]. L3 and U3 are also determined based on in-depth analysis of extensive operational monitoring data from axial field motors and records related to magnetic pole offset faults. This data covers the motor's operating status under different operating conditions and the corresponding magnetic field vibration synchronization anomaly index.

[0175] Furthermore, when MVo is greater than U3 or less than L3, a magnetic pole shift fault type is determined. A third confidence score, C3, is then calculated, based on the magnitude of the excess. Let ΔMVo be the magnitude of the excess. If MVo is greater than U3, ΔMVo equals MVo minus U3; if MVo is less than L3, ΔMVo equals L3 minus MVo. C3 is calculated using function f3, which is determined similarly to the previous method and is based on the correlation between the magnitude of the magnetic field vibration synchronization anomaly index exceeding this range in historical data and the actual probability of a magnetic pole shift fault.

[0176] It can be explained that this function may be different from f1 and f2 in form, but they are all intended to accurately reflect the relationship between the excess amplitude and the possibility of fault, so that the larger the excess amplitude, the higher the value of C3, which indicates that the possibility of magnetic pole shift fault is greater.

[0177] In step S134, the rotor imbalance fault type, the bearing overheat fault type, and the magnetic pole offset fault type are combined to generate the fault type set of the target rotor assembly with the fault, and the first confidence score, the second confidence score, and the third confidence score are combined to generate the fault confidence score of the target rotor assembly with the fault.

[0178] In the disclosed embodiment, the determined rotor imbalance fault type, bearing overheat fault type, and magnetic pole offset fault type are integrated to form a fault type set F. This set F comprehensively includes the possible fault types determined based on the matching of the aforementioned abnormality indices with the thresholds.

[0179] At the same time, the first confidence score C1, the second confidence score C2, and the third confidence score C3 are combined to generate the fault confidence score C. The combination method here can be understood as a specific data combination method, such as arranging C1, C2, and C3 in a certain order or performing some weighting processing to form a comprehensive value C.

[0180] The specific weighting method can be determined based on the importance of different fault types in their actual operation. For example, if rotor imbalance significantly impacts motor stability, C1 might be given a higher weight when determining C. If bearing overheating, while infrequent, has serious consequences if it does occur, the weight of C2 can be adjusted accordingly. This approach allows the fault confidence score C to more accurately reflect the overall likelihood of motor failure and the weights assigned to each fault type.

[0181] After determining the fault type set F and the fault confidence score C, this information needs to be presented in an intuitive and easy-to-understand manner. This requires generating a visual inspection report. First, design the report format and layout. The report's beginning can describe basic information about the axial-field motor rotor assembly being inspected, such as the motor model, the equipment it belongs to, and the inspection date.

[0182] Next, the report's main body details the contents of the fault type set F. Each fault type is described concisely and clearly. For example, for a rotor imbalance fault, describe the potential causes, such as uneven rotor quality during manufacturing or component wear from long-term operation. Also, provide relevant test data as support. This data could be the vibration-temperature coupling anomaly index VTa calculated previously and its comparison with the first fault threshold range.

[0183] For bearing overheating faults, the system also explains possible causes, such as insufficient lubrication or bearing quality issues, and displays the matching information between the temperature-magnetic field interaction anomaly index TMo and the second fault threshold range. For magnetic pole offset faults, it explains possible causes such as improper installation or external magnetic field interference, and compares the magnetic field vibration synchronization anomaly index MVo with the third fault threshold range.

[0184] After presenting the fault types, focus on presenting the fault confidence score C. Explain how C is derived by combining C1, C2, and C3, and the significance of each score. For example, explain that C1 represents the confidence level for a rotor imbalance fault, C2 for a bearing overheating fault, and C3 for a magnetic pole misalignment fault, while C comprehensively reflects the overall probability of the fault.

[0185] To make the report more intuitive, use visualization elements such as charts. For example, you can create a bar chart with the horizontal axis representing different fault types and the vertical axis representing the corresponding confidence scores. This allows you to clearly see the likelihood of each fault type occurring. Alternatively, use a radar chart, using different fault types as dimensions and the fault confidence score as the radius, to display the overall fault situation.

[0186] After the visual inspection report is generated, it is transmitted to the target terminal for display. The target terminal can be a computer screen in the factory control room or a mobile device held by maintenance personnel. The transmission process adopts a suitable network communication protocol, such as a network transmission method based on the TCP / IP protocol. First, on the device side that generates the report, the report data is encapsulated according to the selected protocol, and necessary header information is added, such as the IP address and port number of the target terminal. The encapsulated data is then sent out through the network interface. On the target terminal side, the corresponding port is listened to, and after receiving the data, it is decapsulated according to the same protocol, the report data is extracted, and displayed on the terminal display interface, so that relevant personnel can obtain the fault information of the axial magnetic field motor rotor assembly in a timely manner and take corresponding maintenance measures.

[0187] The present disclosure also provides an axial magnetic field motor rotor assembly fault detection system, see Figure 2 As shown, the system includes:

[0188] a construction module 210 configured to collect vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor in an operating state, and construct a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data;

[0189] a first generating module 220 configured to perform dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and perform multi-dimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly;

[0190] A determination module 230 is configured to determine a set of fault types and corresponding fault confidence scores of the target rotor assembly having a fault based on a matching result between the set of comprehensive abnormality indicators and a preset fault determination threshold;

[0191] The second generating module 240 is configured to generate a visual inspection report for the target rotor assembly according to the fault type set and the corresponding fault confidence score of the target rotor assembly, and the visual inspection report is used to display the fault result on the terminal.

[0192] In a possible implementation, the first generating module 220 is configured to:

[0193] Performing frequency domain transformation processing on the vibration waveform data to generate the vibration coupling feature;

[0194] Performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and using the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature;

[0195] Performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve, and determining the magnetic field offset characteristics according to the curvature change points of the gradient curve;

[0196] The vibration coupling characteristics, temperature abnormal fluctuation characteristics and magnetic field offset characteristics are respectively subjected to sliding window normalization processing, and the normalized characteristics are associated according to the time-space dimension to generate the dynamic operation feature set containing time series and space correlation.

[0197] In a possible implementation, the first generating module 220 is configured to:

[0198] Dividing the vibration waveform data into a plurality of time series segments of equal length, performing fast Fourier transform processing on each of the time series segments to generate corresponding vibration spectrum segments;

[0199] Extract the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of each vibration spectrum segment;

[0200] Performing sliding average processing on the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of multiple time series segments to generate a continuous vibration spectrum feature;

[0201] Setting a first target frequency band corresponding to the rotor structural resonance frequency of the rotor assembly, and calculating a first proportional coefficient of the integral value of the vibration energy in the first target frequency band to the integral value of the vibration energy in the entire frequency band;

[0202] Setting a second target frequency band corresponding to the characteristic frequency of bearing wear of the rotor assembly, and calculating a second proportional coefficient of the integral value of the vibration energy in the second target frequency band to the integral value of the vibration energy in the entire frequency band;

[0203] Setting a weight distribution ratio of the first proportional coefficient to the second proportional coefficient according to a correlation between historical fault data of the rotor structure resonance frequency and the bearing wear characteristic frequency;

[0204] The sum of the weighted first proportional coefficient and the second proportional coefficient is used as the vibration coupling feature.

[0205] In a possible implementation, the first generating module 220 is configured to:

[0206] Setting the length of the sliding window to a preset time period, performing a linear regression analysis on the temperature data of each temperature sensing unit within the sliding window to obtain a temperature change trend equation;

[0207] extracting a slope parameter from the temperature change trend equation as a temperature change trend slope corresponding to the temperature sensing unit;

[0208] When the temperature change trend slope exceeds a preset positive change threshold, the temperature sensing unit is marked as an abnormal temperature rising unit;

[0209] Acquiring a physical distance parameter between adjacent temperature sensing units, and calculating a temperature conduction delay compensation coefficient based on the physical distance parameter;

[0210] Performing time alignment processing on the temperature change trend slopes of adjacent temperature sensing units according to the temperature conduction delay compensation coefficient;

[0211] The absolute value of the slope difference between the aligned adjacent temperature sensing units is calculated, and the absolute value of the difference is compared with a preset difference threshold to generate the abnormal temperature fluctuation feature.

[0212] In a possible implementation, the first generating module 220 is configured to:

[0213] Constructing a three-dimensional distribution surface of magnetic field intensity according to the spatial coordinate positions of multiple magnetic field detection devices;

[0214] Slicing the three-dimensional distribution surface along the axial direction of the rotor assembly to extract magnetic field intensity distribution data of each slice plane;

[0215] Perform polynomial fitting on the magnetic field intensity distribution data of each slice plane to generate the corresponding axial magnetic field intensity gradient curve;

[0216] The similarity coefficients between the gradient curves of adjacent slice planes are calculated, and the slice planes with similarity coefficients lower than a preset threshold are marked as magnetic field anomaly areas.

[0217] In a possible implementation, the building module 210 is configured as follows:

[0218] Deploying a plurality of vibration sensors at preset positions of the rotor assembly, and collecting the vibration waveform data at a first sampling period through the vibration sensors, wherein the vibration waveform data includes time-series waveforms of an axial vibration component and a radial vibration component;

[0219] A plurality of temperature sensing units are provided on the surface of the rotor assembly, and the temperature gradient distribution data is collected by the temperature sensing units at a second sampling period, wherein the temperature gradient distribution data includes temperature change rates of different regions and temperature differences between adjacent regions;

[0220] a plurality of magnetic field detection devices are provided along the circumferential direction of the rotor assembly, and the magnetic field intensity distribution data is collected by the magnetic field detection devices at a third sampling period, wherein the magnetic field intensity distribution data includes an axial magnetic field intensity distribution curve and a magnetic field asymmetry coefficient;

[0221] Performing time stamp synchronization processing on the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data, respectively, and unifying data with inconsistent sampling frequencies to the same time base using interpolation or resampling methods;

[0222] A data mapping table is established according to the spatial position relationship of the sensors, and the synchronized data are fused in a time-space alignment manner to generate the dynamic monitoring data set.

[0223] In a possible implementation, the first generating module 220 is configured to:

[0224] Establishing a first correlation mapping relationship between the vibration coupling feature and the temperature abnormal fluctuation feature, and calculating a vibration-temperature coupling abnormality index based on the first correlation mapping relationship;

[0225] Establishing a second correlation mapping relationship between the temperature anomaly fluctuation characteristics and the magnetic field offset characteristics, and calculating a temperature-magnetic field interaction anomaly index based on the second correlation mapping relationship;

[0226] establishing a third association mapping relationship between the magnetic field offset feature and the vibration coupling feature, and calculating a magnetic field vibration synchronization anomaly index based on the third association mapping relationship;

[0227] The vibration-temperature coupling anomaly index, temperature-magnetic field interaction anomaly index and magnetic field-vibration synchronization anomaly index are normalized respectively, and dynamic weight coefficients are allocated according to the contribution of each type of anomaly index in the historical fault data. The normalized anomaly indexes are weighted and fused based on the dynamic weight coefficients to generate the comprehensive anomaly indicator set.

[0228] In a possible implementation, the determining module 230 is configured to:

[0229] Matching the vibration-temperature coupling anomaly index with a first fault threshold interval; if the matching result is beyond the first fault threshold interval, determining that a rotor imbalance fault type exists, and calculating a first confidence score based on the excess magnitude;

[0230] Matching the temperature-magnetic field interaction anomaly index with a second fault threshold interval; if the matching result is beyond the second fault threshold interval, determining that a bearing overheating fault type exists, and calculating a second confidence score based on the excess magnitude;

[0231] Matching the magnetic field vibration synchronization anomaly index with a third fault threshold interval; if the matching result is beyond the third fault threshold interval, determining that a magnetic pole shift fault type exists, and calculating a third confidence score based on the excess magnitude;

[0232] The rotor imbalance fault type, bearing overheat fault type and magnetic pole offset fault type are combined to generate the fault type set of the target rotor assembly with the fault, and the first confidence score, the second confidence score and the third confidence score are combined to generate the fault confidence score of the target rotor assembly with the fault.

[0233] The present disclosure also provides an electronic device, including:

[0234] A memory having a computer program stored thereon; a processor configured to execute the computer program in the memory to implement the steps of the method described in any one of the aforementioned embodiments.

[0235] Figure 3 The axial magnetic field motor rotor assembly fault detection device 100 shown includes: a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, such as through a bus 1002. Optionally, the axial magnetic field motor rotor assembly fault detection device 100 may also include a communication component 1004, which can be used for data interaction between the device 100 and other devices, such as data transmission and / or data reception. It should be noted that in actual scheduling, the communication component 1004 is not limited to one, and the structure of the axial magnetic field motor rotor assembly fault detection device 100 does not constitute a limitation on the embodiments of the present application.

[0236] Processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 1001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, or a combination of a DSP and a microprocessor.

[0237] Bus 1002 may include a path for transmitting information between the above components. Bus 1002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0238] The memory 1003 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store program code and can be read by a computer, without limitation herein.

[0239] The memory 1003 is used to store program codes for executing the embodiments of the present disclosure, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the embodiment of the axial magnetic field motor rotor assembly fault detection method.

[0240] An embodiment of the present disclosure also provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps and corresponding contents of the aforementioned embodiment of the axial magnetic field motor rotor assembly fault detection method can be implemented.

[0241] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, various changes, modifications, replacements and variations can be made to these embodiments, and these changes, modifications, replacements and variations all fall within the scope of protection of the present disclosure.

[0242] It should also be noted that the various specific technical features described in the above specific embodiments may be combined in any suitable manner, unless there is any contradiction, and these combinations shall also be considered as the contents disclosed in this disclosure. To avoid unnecessary repetition, this disclosure will not further describe various possible combinations. The technical scope of this application is not limited to the contents of the specification and must be determined based on the scope of the claims.

Claims

1. A method for detecting faults in an axial magnetic field motor rotor assembly, characterized in that: The method comprises: collecting vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor when in operation, and constructing a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data; performing dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and performing multidimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly; Determining a fault type set and a corresponding fault confidence score of the target rotor assembly having a fault based on a matching result between the comprehensive abnormality indicator set and a preset fault determination threshold; Generate a visual inspection report for the target rotor assembly according to the fault type set and the corresponding fault confidence score of the target rotor assembly, wherein the visual inspection report is used to display the fault result on the terminal; The multi-dimensional correlation mapping process is performed based on the dynamic operation feature set to generate a comprehensive abnormality indicator set of the rotor assembly, including: Establishing a first correlation mapping relationship between the vibration coupling feature and the temperature abnormal fluctuation feature, and calculating a vibration-temperature coupling abnormality index based on the first correlation mapping relationship; Establishing a second correlation mapping relationship between the temperature anomaly fluctuation characteristics and the magnetic field offset characteristics, and calculating a temperature-magnetic field interaction anomaly index based on the second correlation mapping relationship; establishing a third association mapping relationship between the magnetic field offset feature and the vibration coupling feature, and calculating a magnetic field vibration synchronization anomaly index based on the third association mapping relationship; The vibration-temperature coupling anomaly index, temperature-magnetic field interaction anomaly index and magnetic field-vibration synchronization anomaly index are normalized respectively, and dynamic weight coefficients are allocated according to the contribution of each type of anomaly index in the historical fault data. The normalized anomaly indexes are weighted and fused based on the dynamic weight coefficients to generate the comprehensive anomaly indicator set.

2. The method for detecting faults in an axial magnetic field motor rotor assembly according to claim 1, characterized in that: The dynamic feature extraction processing is performed on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features and magnetic field offset features, including: Performing frequency domain transformation processing on the vibration waveform data to generate the vibration coupling feature; Performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and using the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature; Performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve, and determining the magnetic field offset characteristics according to the curvature change points of the gradient curve; The vibration coupling characteristics, temperature abnormal fluctuation characteristics and magnetic field offset characteristics are respectively subjected to sliding window normalization processing, and the normalized characteristics are associated according to the time-space dimension to generate the dynamic operation feature set containing time series and space correlation.

3. The method for detecting faults in an axial magnetic field motor rotor assembly according to claim 2, characterized in that: The performing frequency domain transformation processing on the vibration waveform data to generate the vibration coupling feature includes: Dividing the vibration waveform data into a plurality of time series segments of equal length, performing fast Fourier transform processing on each of the time series segments to generate corresponding vibration spectrum segments; Extract the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of each vibration spectrum segment; Performing sliding average processing on the fundamental frequency component amplitude, harmonic component amplitude and high-frequency noise energy value of multiple time series segments to generate a continuous vibration spectrum feature; Setting a first target frequency band corresponding to the rotor structural resonance frequency of the rotor assembly, and calculating a first proportional coefficient of the integral value of the vibration energy in the first target frequency band to the integral value of the vibration energy in the entire frequency band; Setting a second target frequency band corresponding to the characteristic frequency of bearing wear of the rotor assembly, and calculating a second proportional coefficient of the integral value of the vibration energy in the second target frequency band to the integral value of the vibration energy in the entire frequency band; Setting a weight distribution ratio of the first proportional coefficient to the second proportional coefficient according to a correlation between historical fault data of the rotor structure resonance frequency and the bearing wear characteristic frequency; The sum of the weighted first proportional coefficient and the second proportional coefficient is used as the vibration coupling feature.

4. The method for detecting faults of an axial magnetic field motor rotor assembly according to claim 2, wherein: The performing sliding window statistical analysis on the temperature gradient distribution data, calculating the temperature change trend slope corresponding to each temperature sensing unit, and taking the difference between the temperature change trend slopes of adjacent temperature sensing units as the temperature abnormal fluctuation feature, includes: Setting the length of the sliding window to a preset time period, performing a linear regression analysis on the temperature data of each temperature sensing unit within the sliding window to obtain a temperature change trend equation; extracting a slope parameter from the temperature change trend equation as a temperature change trend slope corresponding to the temperature sensing unit; When the temperature change trend slope exceeds a preset positive change threshold, the temperature sensing unit is marked as an abnormal temperature rising unit; Acquiring a physical distance parameter between adjacent temperature sensing units, and calculating a temperature conduction delay compensation coefficient based on the physical distance parameter; Performing time alignment processing on the temperature change trend slopes of adjacent temperature sensing units according to the temperature conduction delay compensation coefficient; The absolute value of the slope difference between the aligned adjacent temperature sensing units is calculated, and the absolute value of the difference is compared with a preset difference threshold to generate the abnormal temperature fluctuation feature.

5. The method for detecting faults of an axial magnetic field motor rotor assembly according to claim 2, characterized in that: The performing spatial distribution fitting on the magnetic field intensity distribution data to generate an axial magnetic field intensity gradient curve includes: Constructing a three-dimensional distribution surface of magnetic field intensity according to the spatial coordinate positions of multiple magnetic field detection devices; Slicing the three-dimensional distribution surface along the axial direction of the rotor assembly to extract magnetic field intensity distribution data of each slice plane; Perform polynomial fitting on the magnetic field intensity distribution data of each slice plane to generate the corresponding axial magnetic field intensity gradient curve; The similarity coefficients between the gradient curves of adjacent slice planes are calculated, and the slice planes with similarity coefficients lower than a preset threshold are marked as magnetic field anomaly areas.

6. The method for detecting faults of an axial magnetic field motor rotor assembly according to claim 1, characterized in that: The collecting of vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor in the operating state, and constructing a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data, includes: Deploying a plurality of vibration sensors at preset positions of the rotor assembly, and collecting the vibration waveform data at a first sampling period through the vibration sensors, wherein the vibration waveform data includes time-series waveforms of an axial vibration component and a radial vibration component; A plurality of temperature sensing units are provided on the surface of the rotor assembly, and the temperature gradient distribution data is collected by the temperature sensing units at a second sampling period, wherein the temperature gradient distribution data includes temperature change rates of different regions and temperature differences between adjacent regions; a plurality of magnetic field detection devices are provided along the circumferential direction of the rotor assembly, and the magnetic field intensity distribution data is collected by the magnetic field detection devices at a third sampling period, wherein the magnetic field intensity distribution data includes an axial magnetic field intensity distribution curve and a magnetic field asymmetry coefficient; Performing time stamp synchronization processing on the vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data, respectively, and unifying data with inconsistent sampling frequencies to the same time base using interpolation or resampling methods; A data mapping table is established according to the spatial position relationship of the sensors, and the synchronized data are fused in a time-space alignment manner to generate the dynamic monitoring data set.

7. The method for detecting faults in an axial magnetic field motor rotor assembly according to claim 1, characterized in that: The determining of the fault type set and the corresponding fault confidence score of the target rotor assembly having a fault based on the matching result between the comprehensive abnormality indicator set and the preset fault judgment threshold value includes: Matching the vibration-temperature coupling anomaly index with a first fault threshold interval; if the matching result is beyond the first fault threshold interval, determining that a rotor imbalance fault type exists, and calculating a first confidence score based on the excess magnitude; Matching the temperature-magnetic field interaction anomaly index with a second fault threshold interval; if the matching result is beyond the second fault threshold interval, determining that a bearing overheating fault type exists, and calculating a second confidence score based on the excess magnitude; Matching the magnetic field vibration synchronization anomaly index with a third fault threshold interval; if the matching result is beyond the third fault threshold interval, determining that a magnetic pole shift fault type exists, and calculating a third confidence score based on the excess magnitude; The rotor imbalance fault type, bearing overheat fault type and magnetic pole offset fault type are combined to generate the fault type set of the target rotor assembly with the fault, and the first confidence score, the second confidence score and the third confidence score are combined to generate the fault confidence score of the target rotor assembly with the fault.

8. A fault detection system for an axial magnetic field motor rotor assembly, characterized in that: The system comprises: a construction module configured to collect vibration waveform data, temperature gradient distribution data, and magnetic field intensity distribution data of the rotor assembly in the axial magnetic field motor in an operating state, and construct a dynamic monitoring data set based on the vibration waveform data, the temperature gradient distribution data, and the magnetic field intensity distribution data; a first generating module configured to perform dynamic feature extraction processing on the data in the dynamic monitoring data set to generate a dynamic operation feature set including vibration coupling features, abnormal temperature fluctuation features, and magnetic field offset features, and perform multi-dimensional correlation mapping processing based on the dynamic operation feature set to generate a comprehensive abnormality indicator set for the rotor assembly; a determination module configured to determine a set of fault types and corresponding fault confidence scores of the target rotor assembly having a fault based on a matching result between the set of comprehensive abnormality indicators and a preset fault determination threshold; a second generating module configured to generate a visual inspection report for the target rotor assembly based on the fault type set and the corresponding fault confidence score of the target rotor assembly, wherein the visual inspection report is used to display the fault result on the terminal; The first generating module is configured to: establish a first correlation mapping relationship between the vibration coupling feature and the temperature abnormal fluctuation feature, and calculate a vibration-temperature coupling abnormality index based on the first correlation mapping relationship; Establishing a second correlation mapping relationship between the temperature anomaly fluctuation characteristics and the magnetic field offset characteristics, and calculating a temperature-magnetic field interaction anomaly index based on the second correlation mapping relationship; establishing a third association mapping relationship between the magnetic field offset feature and the vibration coupling feature, and calculating a magnetic field vibration synchronization anomaly index based on the third association mapping relationship; The vibration-temperature coupling anomaly index, temperature-magnetic field interaction anomaly index and magnetic field-vibration synchronization anomaly index are normalized respectively, and dynamic weight coefficients are allocated according to the contribution of each type of anomaly index in the historical fault data. The normalized anomaly indexes are weighted and fused based on the dynamic weight coefficients to generate the comprehensive anomaly indicator set.

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 7.

Images