A method for detecting the center of magnetic field of a motor

Abstract

The present invention belongs to the technical field of motor magnetic field center detection, and specifically discloses a motor magnetic field center detection method, which realizes static and dynamic contactless measurement of the motor magnetic field center by utilizing magnetic imaging technology in conjunction with vibration synchronous detection. The static detection specifically obtains a static magneto-optical image when the motor is stationary, extracts the magnetic field center and at least three reference stator positions, and realizes non-contact static magnetic field center detection based on multiple reference points by comparing the geometric offset between the magnetic field center and these reference points. The dynamic detection specifically adopts magnetic imaging technology to synchronously capture magneto-optical images and motor vibration data under the dynamic increasing state of motor speed, so as to realize real-time detection of the motor under continuous operation. In addition, the magnetic field center offset distance extracted from the magneto-optical image is correlated and corrected by synchronous motor vibration data, which can minimize the influence of vibration on the detection result and greatly improve the detection accuracy and reliability.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor magnetic field center detection, and specifically discloses a motor magnetic field center detection method. Background Art

[0002] As a power component in modern industry and everyday life, motors primarily utilize the interaction between electric current and magnetic fields to achieve rotation, meeting the power requirements of various application scenarios. The centerline or equilibrium point of the motor's internal magnetic field during operation is called the magnetic field center, which is crucial for ensuring symmetry and efficiency. Any deviation from the magnetic field center can cause vibration, noise, and reduced efficiency. Therefore, accurately detecting the magnetic field center is essential for ensuring motor performance.

[0003] Solutions for detecting the motor's magnetic field center are already available. For example, Chinese Invention Patent Publication No. CN117232370A proposes a magnetic field center detection tool. This tool utilizes a measuring rod, a positioning rod, and a sliding measuring assembly. Scale lines and an LED light on the measuring rod assist in aligning the motor's magnetic field center. This tool is primarily used to quickly measure the relative position difference between the stator and rotor under static conditions to ensure magnetic field center alignment during motor installation.

[0004] This method, based on a physical measuring rod, is primarily suitable for detection in static environments because it relies on mechanical contact measurement. Under static conditions, when the motor is not rotating and the stator and rotor positions are fixed, this type of contact measurement is suitable. However, during motor operation, due to the dynamic changes in actual operating conditions, the magnetic field center may also shift. Therefore, detecting the magnetic field center in dynamic environments is also crucial.

[0005] Using the above-mentioned static detection tools in a dynamic environment will face the following limitations: 1. Inability to monitor in real time: Static tools cannot provide continuous feedback on the magnetic field center position while the motor is running.

[0006] 2. Poor adaptability: Static measurement results may no longer be accurate due to changes in internal conditions when the motor is running.

[0007] 3. Operational limitations: Contact measurement is difficult to implement in dynamic environments and may lead to safety hazards or damage to equipment.

[0008] These limitations make it impossible to effectively monitor the magnetic field center during the actual operation of the motor.

[0009] Another example is the central magnetic field automatic tester proposed in Chinese invention patent publication number CN104035057A, which detects the central magnetic field on the surface of the magnetic steel through a motor-driven rotating structure and a Hall probe.

[0010] This solution uses non-contact sensor technology to detect the magnetic field during the intermittent period when the motor stops rotating. Although it can measure the magnetic field center under dynamic conditions to a certain extent, it can only detect when the motor is stopped. It cannot achieve real-time monitoring under conditions of continuous motor operation or different speeds. This limits its ability to capture dynamic changes in the magnetic field center during motor operation, which may lead to the omission of critical offset information. In addition, the vibration generated during motor operation will have a certain impact on the position of the magnetic field center, such as causing magnetic circuit asymmetry, which will cause the magnetic field center to deviate. However, this solution does not consider the interference of vibration on the detection results, which may cause distortion of the detection data and reduce the detection accuracy and reliability. Summary of the Invention

[0011] To this end, one purpose of an embodiment of the present application is to provide a method for detecting the magnetic field center of a motor, which realizes static and dynamic contactless measurement of the magnetic field center of the motor and real-time, high-precision detection requirements of the motor in a continuous operation state by utilizing magnetic imaging technology in combination with vibration synchronous detection, thereby solving the problems mentioned in the background technology.

[0012] The purpose of the present invention can be achieved through the following technical solutions: A method for detecting the magnetic field center of a motor, comprising the following steps: S1. Static detection: S11. Fix the motor and expose the stator core and rotor poles, while aligning the magnetic imaging terminal axially with the motor poles.

[0013] S12. Use the magnetic imaging terminal to collect static magneto-optical images and extract the magnetic field center position and at least three reference stator positions.

[0014] S13. Analyze the static magnetic field center offset by comparing the geometric offsets between the magnetic field center and each reference stator position.

[0015] S2. Dynamic detection: S21. Start the motor and dynamically increase the speed to the target speed in a preset speed step size, synchronously capturing the magneto-optical image sequence and motor vibration data sequence at each speed point.

[0016] S22. Perform ghost recognition on the magneto-optical image sequence at each speed point, and select qualified magneto-optical images and synchronous motor vibration data accordingly.

[0017] S23. Extract the magnetic field center position from the qualified magneto-optical image corresponding to each speed point and compare it with the same reference stator position to calculate the dynamic geometric offset, and perform correlation analysis with the synchronous motor vibration data. Based on this, perform vibration correction on the magnetic field center offset under dynamic detection.

[0018] S3. Detection fusion: The magnetic field center offset after static and dynamic detection correction is integrated and weighted fusion is used to generate the final offset.

[0019] Combining all the above technical solutions, the positive effects of the present invention are as follows: 1. The present invention uses magnetic imaging technology to obtain static magneto-optical images when the motor is stationary, extracts the magnetic field center and at least three reference stator positions, and compares the geometric offset between the magnetic field center and these reference points to realize non-contact static magnetic field center detection based on multiple reference points. On the one hand, comparative analysis through multiple reference points can more accurately determine the position of the magnetic field center and its offset, thereby improving the detection accuracy. On the other hand, it avoids the mechanical wear and installation errors that may be caused by traditional contact measurement tools, thereby ensuring the safety and reliability of the detection process.

[0020] 2. The present invention adopts magnetic imaging technology to synchronously capture magneto-optical images and motor vibration data when the motor speed is dynamically increasing, which can realize non-contact magnetic field center detection when the motor is in continuous operation, and provide more realistic dynamic magnetic field center data. In addition, the magnetic field center offset distance extracted from the magneto-optical image is correlated and corrected by synchronous motor vibration data, which can minimize the impact of vibration on the detection results and greatly improve the detection accuracy and reliability.

[0021] 3. The present invention combines the magnetic field center offsets after static and dynamic detection corrections to generate the final magnetic field center offset, ensuring a comprehensive and accurate magnetic field center evaluation. It can not only capture the magnetic field center offset characteristics of the motor under different working conditions, but also effectively reduce the errors that may be caused by a single detection method, thereby improving the accuracy and reliability of the overall detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0023] Figure 1 This is a diagram of the steps for implementing the method of the present invention.

[0024] Figure 2 It is a schematic diagram of the magnetic imaging terminal of the present invention.

[0025] Figure 3 This is a flowchart of the implementation of starting the motor and dynamically increasing the speed to the target speed with a preset speed step size in the present invention. DETAILED DESCRIPTION

[0026] 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.

[0027] The magnetic imaging technology used in the present invention is based on the Faraday effect or the Kerr effect, and uses the rotation characteristics of polarized light in a magnetic field to visualize the magnetic field distribution. The magneto-optical image generated by this technology can intuitively display the magnetic field distribution, thereby assisting in accurately locating the magnetic field center offset.

[0028] See also Figure 1 As shown, the present invention proposes a method for detecting the magnetic field center of a motor, comprising the following steps: S1. Static detection: S11. Fix the motor and expose the stator core and rotor poles, while aligning the magnetic imaging terminal with the axial direction of the motor poles.

[0029] It should be added that before static testing, you first need to select a stable experimental environment to prevent external factors such as electromagnetic interference and temperature from affecting the measurement results; then fix the motor to ensure that the motor does not move or vibrate during the test; finally, you need to remove part of the motor casing to expose the stator core and rotor poles so that the magnetic imaging terminal can directly image them.

[0030] See also Figure 2 As shown in Figure 1, the aforementioned magnetic imaging terminal includes a magneto-optical film, a polarized light source, and a high-resolution CCD camera. The magneto-optical film is applied to the surface of the stator core or rotor poles to enhance the contrast of the magnetic field visualization. It operates based on the Faraday effect or the Kerr effect. When subjected to an external magnetic field, its optical properties change, such as the rotation angle of polarized light, allowing the magnetic field distribution to be visually displayed optically.

[0031] The polarized light source provides a precisely controlled polarized light beam to illuminate the magneto-optical film. Since the polarized light will rotate at different angles when passing through the magneto-optical film with different magnetic field strength areas, the change in the magnetic field can be converted into a change in light intensity or phase, which facilitates subsequent image acquisition and analysis.

[0032] A high-resolution CCD camera is used to capture polarized light images modulated by the magneto-optical film. This camera records the magnetic field distribution at high resolution, generating clear magneto-optical images. These images provide detailed magnetic field information, including the location and offset of the magnetic field center, providing the data foundation for subsequent magnetic field center positioning and correction.

[0033] Through the collaborative work of these three components, magnetic imaging technology can accurately and intuitively display the magnetic field distribution inside the motor without contact, helping to accurately locate the magnetic field center and assess its offset. This method not only improves detection accuracy but also avoids the mechanical damage and errors that may be caused by traditional contact measurement.

[0034] It should be understood that the magnetic imaging terminal needs to cover key areas of the motor stator and rotor to fully record the magnetic field distribution. If the magnetic imaging terminal is not axially aligned with the motor poles, the magnetic field information in these areas may not be fully captured, resulting in data loss or incompleteness. Therefore, the magnetic imaging terminal is axially aligned with the motor poles before static testing. Axial alignment can reduce blind spots caused by angular deviations and ensure that all important magnetic field features can be clearly recorded. At the same time, it can keep the working plane of the magnetic imaging terminal parallel to the surface of the motor poles, thereby maximizing the resolution and clarity of the magneto-optical image, facilitating subsequent image processing and analysis. More importantly, the magnetic imaging terminal is very sensitive to the direction of the magnetic field. If the terminal is not axially aligned with the motor poles, it may cause deviations in the measurement results of the magnetic field strength and direction, which in turn affects the calculation of the center position of the magnetic field.

[0035] S12. Use the magnetic imaging terminal to collect static magneto-optical images and extract the magnetic field center position and at least three reference stator positions.

[0036] In the preferred embodiment of the above scheme, the specific implementation steps of extracting the magnetic field center position from the static magneto-optical image are as follows: preprocessing the collected static magneto-optical image, including removing noise and enhancing contrast, to improve the accuracy of subsequent processing.

[0037] Image segmentation is used to separate the magnetic field distribution area from the background in static magneto-optical images.

[0038] The spatial distribution of magnetic field intensity is generated based on the polarization characteristics of the magneto-optical film and grayscale or color information. Usually, the magnetic field intensity is represented as grayscale or color changes in the image.

[0039] The magnetic field intensity gradient in the image is used to determine the boundaries of the magnetic poles, and the magnetic field center is calculated from the geometric center.

[0040] In a further preferred embodiment of the above solution, the specific extraction process of at least three reference stator positions is as follows: a closed boundary area is defined in the static magneto-optical image according to the actual positions of the plurality of stators of the motor.

[0041] The geometric center point is calculated in this closed boundary area, and starting from the center point, multiple directional rays are diverged toward the boundary line at uniform intervals along the radial direction to form multiple directional rays.

[0042] The stator position corresponding to the point where each directional ray intersects the boundary line is regarded as the reference stator position.

[0043] In the specific operation of the above embodiment, if the goal is to extract four reference stator positions, the system should start from the geometric center point of the closed boundary region and diverge outward along the radial direction at angular intervals of 90 degrees (i.e., 360 degrees divided by 4), thereby forming four directional rays. Each ray will intersect the boundary at a point, and the stator positions corresponding to these four intersection points will serve as the reference stator positions.

[0044] It should be pointed out that the main purpose of selecting the reference stator position is to provide a reliable reference framework, and the purpose of selecting multiple reference stator positions is to effectively reduce the impact of noise or local anomalies that may be introduced by a single data point on the results, thereby improving the stability and reliability of the overall detection.

[0045] It is also worth noting that when selecting multiple reference stator positions, the method of selecting reference points diverging from the center point to the boundary line ensures that these points are evenly distributed within the stator boundary area. This uniform distribution helps to fully cover the magnetic field distribution of the motor and avoids ignoring information in certain local areas.

[0046] S13. Analyze the static magnetic field center offset by comparing the geometric offsets between the magnetic field center and each reference stator position.

[0047] The specific analysis process is as follows: the geometric offsets of the magnetic field center and each reference stator position are compared, the maximum offset and the average offset are selected, and the standard deviation of the geometric offsets of all reference stator positions is calculated and recorded as the static offset standard deviation.

[0048] It is important to note that the geometric offset mentioned above refers to the difference in distance between the magnetic field center and each reference stator position. Specifically, it is determined by calculating the straight-line distance between the magnetic field center and each selected reference stator position.

[0049] The maximum offset and average offset mentioned above take multiple dimensions into consideration, among which the maximum offset reflects the maximum deviation of the magnetic field center relative to the reference stator position, which helps to identify the most serious offset situation.

[0050] The average offset provides a quantitative description of the overall offset and can reflect the general state of the entire magnetic field distribution.

[0051] The static offset standard deviation measures the degree of dispersion between the reference stator offsets.

[0052] The static offset standard deviation is normalized so that its range falls within [0, 1].

[0053] The normalized static offset standard deviation is used to assign weights to the maximum offset and the average offset.

[0054] The specific operation of the above weight assignment is to use the normalized offset standard deviation as a weight adjustment factor. When the static offset standard deviation is small, that is, the offset is relatively uniform, more attention is paid to the average offset; when the static offset standard deviation is large, that is, there is a large dispersion, more attention is paid to the maximum offset. This mechanism can adaptively adjust the evaluation focus according to the characteristics of the actual magnetic field distribution, ensuring more scientific and reasonable evaluation results.

[0055] Exemplarily, the normalized static offset standard deviation is used as the weight value of the maximum offset, and the difference between the value 1 and the normalized static offset standard deviation is used as the weight value of the average offset.

[0056] The static magnetic field center offset is obtained by combining the maximum offset, the average offset and their corresponding weights for weighted calculation.

[0057] It should be noted that in the motor structure, the stator assembly is fixed on a circular trajectory. In theory, the motor's magnetic field center should be equidistant from every stator position on this circumference. This symmetrical layout ensures balanced and efficient motor operation. By comparing and analyzing the geometric offset between the magnetic field center and each reference stator position, the degree of magnetic field center deviation can be quantified. However, relying solely on the maximum offset or average offset may not fully reflect the true state of the magnetic field distribution. For example, the maximum offset may mask the consistency of the overall distribution, while the average offset may ignore the impact of extreme offsets. By combining these two indicators and introducing the offset standard deviation for dynamic adjustment, a more comprehensive assessment of magnetic field center deviation can be achieved. In addition, this method can not only detect significant offset phenomena but also identify potential problem patterns, such as systematic deviations or random fluctuations, through changes in the offset standard deviation.

[0058] It is important to note that when a motor is static and not running, a magnetic field still exists inside the motor due to the presence of permanent magnets and the residual magnetism effect. This phenomenon provides the necessary foundation for static magnetic field center detection.

[0059] S2. Dynamic detection: S21. Start the motor and dynamically increase the speed to the target speed in a preset speed step size, synchronously capturing the magneto-optical image sequence and motor vibration data sequence at each speed point.

[0060] It should be emphasized that before dynamic testing, a suitable experimental environment needs to be selected to avoid electromagnetic interference, temperature, etc. affecting the test results.

[0061] It should be noted that in dynamic detection of the motor's magnetic field center, in order to achieve real-time monitoring while the motor is running, a high-speed polarization camera is required to replace the high-resolution CCD camera in the magnetic imaging terminal used in static detection. This replacement is due to the high-speed polarization camera's higher frame rate, which allows it to capture multiple frames of images in a short period of time, ensuring that magnetic field changes during motor operation can be recorded in real time. This is crucial for capturing rapidly changing magnetic field distributions. Under dynamic conditions, the magnetic field distribution changes as the motor operates. The high-speed polarization camera can respond quickly to these changes while maintaining image quality, ensuring clear and stable magneto-optical images even in high-speed rotation or vibration environments.

[0062] It should be noted that, during dynamic testing, the magnetic imaging terminal also needs to be aligned with the motor pole axis.

[0063] Among the ways in which the above scheme can be implemented, see Figure 3 As shown, starting the motor and dynamically increasing the speed to the target speed with a preset speed step size includes the following: setting a speed range according to the specification parameters of the motor, and the speed range is composed of a lower limit speed and an upper limit speed, wherein the lower limit speed is usually zero or close to zero, and the upper limit speed is the maximum design speed of the motor.

[0064] The entire speed range is divided into three intervals: low speed zone, medium speed zone and high speed zone.

[0065] The specific division is as follows: the entire speed range is divided into three parts to determine the first and second quantile speed values. Specifically, the first quantile speed value is located at one-third of the entire speed range, and the second quantile speed value is located at two-thirds of the entire speed range.

[0066] The lower speed limit and the first fractional speed constitute a low speed zone.

[0067] The first and second quantile speeds constitute a medium speed zone.

[0068] The second fractional speed and the upper limit speed constitute a high-speed zone.

[0069] This method divides the entire speed range into three intervals, each of which occupies one-third of the total speed range, thereby ensuring that each interval has a uniform data distribution.

[0070] The set ratio of the motor's rated speed is used as the preset speed step.

[0071] For example, the above-mentioned setting ratio may be 5%.

[0072] Different multiplication factors are used to divide discrete speed points according to different speed ranges, as follows: a) In the low-speed area, a number of discrete speed points are divided according to a preset speed step.

[0073] b) In the medium speed zone, a number of discrete speed points are divided according to the multiples of the preset speed step.

[0074] For example, the discrete speed points in the medium speed zone may be divided according to twice the preset speed step size.

[0075] c) In the high-speed area, a number of discrete speed points are divided according to increasing multiples of the preset speed step size.

[0076] For example, the discrete speed points in the high-speed zone may be divided according to three times of the preset speed step length.

[0077] Arrange all the discrete speed points obtained by division in ascending order to form the final speed sequence.

[0078] Start the motor and accelerate to the corresponding speed value according to the discrete speed points arranged above.

[0079] The present invention does not evenly divide the entire speed range into several equidistant speed points when starting the motor to gradually accelerate. Instead, it divides the entire speed range into a low-speed zone, a medium-speed zone, and a high-speed zone, and adopts differentiated step strategies for different speed intervals. The reason for this is that the response of the motor in the low-speed zone is more sensitive, and any slight change may have a significant impact on the overall performance. The use of a smaller step size can ensure the accurate capture of these key details, thereby improving the validity and reliability of the data. As the speed increases, the motor is prone to resonance. By increasing the step size, data redundancy and potential resonance problems caused by over-dense sampling can be effectively avoided. In addition, in the high-speed zone, the dynamic characteristics of the motor tend to be stable. A larger step size can significantly reduce the amount of data and computational complexity without affecting the accuracy of the results, thereby improving processing efficiency.

[0080] In a further possible implementation of the above solution, the synchronous capture of the magneto-optical image sequence and the motor vibration data sequence is implemented as follows: a steady-state detection time window is set at each speed point.

[0081] It's important to note that during dynamic testing, to ensure data stability when collecting magneto-optical images at each motor speed, the motor must be operating in a stable state during data acquisition. This is because data collected in an unstable state may introduce noise or deviation, affecting the accurate measurement of the magnetic field center position and its offset. To this end, a steady-state detection time window is set at each speed point to monitor whether the motor has reached and maintained a stable operating state.

[0082] When the actual speed after starting the motor enters the allowable error range of each speed point, the motor operation indicators of adjacent times are collected within the set steady-state detection time window. Specifically, the motor operation indicators can be current, voltage, etc., and the difference of the motor operation indicators of adjacent times is compared with the set critical difference. If the differences of all the motor operation indicators of multiple consecutive adjacent times are less than or equal to the critical difference, the current speed is locked within a fixed time.

[0083] In the example of the above operation, the allowable error range may be ±1%.

[0084] The above-mentioned steady-state detection is performed within the allowable error range of the actual motor speed entering the speed point, thereby ensuring that the motor runs near the expected speed point.

[0085] As another example, the plurality of consecutive adjacent times may be three consecutive times.

[0086] The above method compares the differences of motor operating indicators at adjacent times within the steady-state monitoring time window, and judges whether a stable state is reached based on the difference maintenance time. This method not only considers the changes in instantaneous values, but also pays attention to trend stability, further enhancing the accuracy of judgment.

[0087] Furthermore, if the difference in all operating indicators for multiple consecutive time points is still less than or equal to the critical difference at the end of the steady-state detection time window, the current speed is locked for a fixed duration and data collection continues. This is done so that even if the ideal steady state cannot be achieved within the steady-state detection time window, the current speed can still be locked for a fixed duration for data collection, avoiding excessive time wasted waiting for steady state and ensuring the consistency and efficiency of the detection process.

[0088] It should be pointed out that the critical difference mentioned above is set artificially based on experience, and is used to evaluate the proximity between the motor operating indicators at adjacent time points, serving as a critical threshold for determining whether the motor has reached a stable state.

[0089] During the locking period of each speed point, magneto-optical image acquisition and vibration detection are performed synchronously on the motor to obtain multiple frames of magneto-optical images and multiple motor vibration data corresponding to each speed point, and the magneto-optical image sequence and motor vibration data sequence are arranged in chronological order.

[0090] The synchronous vibration detection mentioned above uses a vibration sensor to synchronously record the vibration signal of the motor. The motor vibration data obtained by the detection may be vibration amplitude, vibration acceleration, etc.

[0091] The above ensures that the timestamps of the magneto-optical image acquisition and vibration detection are consistent within the locked time, which facilitates subsequent joint analysis.

[0092] Therefore, by setting the steady-state detection time window, the stable operation of the motor at each speed point can be effectively ensured, and high-quality magneto-optical images and vibration data can be obtained under steady-state conditions.

[0093] S22. Perform ghost recognition on the magneto-optical image sequence at each speed point, and select qualified magneto-optical images and synchronous motor vibration data accordingly.

[0094] Preferably, ghost recognition is performed on the magneto-optical image sequence at each rotation speed point by referring to the following process: performing denoising and contrast enhancement processing on the magneto-optical image sequence at each rotation speed point.

[0095] The above steps help reduce the impact of noise on edge extraction and ghost recognition through denoising, and can improve the visibility of magnetic field distribution features in the image through contrast enhancement.

[0096] Edge extraction is performed on the preprocessed magneto-optical image sequence. If multiple edges are extracted from a magneto-optical image, ghosting is identified and the image with multiple edges is marked as a ghosting image.

[0097] It's important to understand that while magneto-optical images are acquired in a stable state at each rotational speed during dynamic testing, providing a reliable foundation for image stability, the possibility of image ghosting due to rotational speed variations cannot be completely eliminated. Ghosting can blur the magnetic field distribution, making it difficult to accurately identify the magnetic field center, leading to positioning errors. Therefore, ghosting detection is necessary for the acquired magneto-optical image sequence to retain high-quality, ghost-free, and qualified images, providing reliable data support for subsequent magnetic field center analysis.

[0098] It's worth noting that multiple edge detection technology is used to identify ghosting in magneto-optical images. This method is based on the principle that ghosting often produces multiple distinct boundaries at the same location. Specifically, by performing edge detection on the magneto-optical image, if a pixel location has more than two significant edge responses, the image is considered to have ghosting.

[0099] The advantage of identifying ghosts in this way is that multiple edge detection can accurately capture subtle changes caused by ghosting, even if these changes are very small. This improves the sensitivity and accuracy of ghosting. In addition, by setting a reasonable threshold and the number of edge responses, such as more than two significant edge responses, it can effectively distinguish between true edges in the actual magnetic field distribution and artifacts caused by ghosting, thereby reducing the false positive rate.

[0100] Further preferably, the content of screening the qualified magneto-optical images and synchronous motor vibration data is as follows: ghost images are eliminated from the magneto-optical image sequence at each speed point. If there are remaining images, they are regarded as qualified magneto-optical images, and the timestamp of the qualified magneto-optical image is recorded. Then, based on this timestamp, the motor vibration data corresponding to the timestamp is extracted from the multiple collected motor vibration data as the synchronous motor vibration data.

[0101] The above method records the timestamp of the qualified magneto-optical image and directly uses the timestamp of the qualified magneto-optical image to select the corresponding vibration data, thereby ensuring that each pair of images and vibration data are collected at the same time, thereby improving the reliability and consistency of the data.

[0102] If there are no remaining images, the magneto-optical image sequence is processed through multi-frame synthesis to generate a qualified magneto-optical image. At this time, the motor vibration data corresponding to the middle timestamp of multiple motor vibration data at the corresponding speed point is selected as the synchronous vibration data.

[0103] When all the original images at a certain speed point are marked as ghosting, at least one qualified image can still be obtained through multi-frame synthesis, thereby ensuring that there is available data at each speed point and avoiding the impact of data loss on the overall analysis results. In this case, choosing the vibration data with the intermediate timestamp as the synchronization data is a reasonable alternative. This is because the intermediate timestamp usually represents a relatively stable operating state, and its vibration data is more representative.

[0104] S24. Extract the magnetic field center position from the qualified magneto-optical image corresponding to each speed point and compare it with the same reference stator position to calculate the dynamic geometric offset, and perform correlation analysis with the synchronous motor vibration data. Based on this, perform vibration correction on the magnetic field center offset under dynamic detection.

[0105] Using the same reference stator position at the same speed point ensures that all calculated geometric offsets are based on the same reference datum, avoiding data inconsistency caused by changes in the reference point. At the same time, by fixing the reference stator position, the geometric offsets at different speed points are highly comparable.

[0106] In the innovative implementation of the above scheme, the correlation analysis process with the synchronous motor vibration data is as follows: for the geometric offset of the magnetic field center at each speed point, a speed-offset curve is drawn in a coordinate system constructed with the speed value corresponding to the speed point as the horizontal axis and the geometric offset as the vertical axis.

[0107] Similarly, the speed-vibration curve is drawn based on the synchronous motor vibration data corresponding to each speed point.

[0108] The slopes of the data points on the speed-offset curve and the speed-vibration curve are calculated respectively, and the points where the absolute value of the slope reaches the configured threshold are marked as change points.

[0109] This method based on slope calculation and threshold setting can effectively identify the locations where the curve changes significantly.

[0110] The speed-vibration curve is divided into a stable segment and a non-stationary segment based on the change points marked on the speed-vibration curve.

[0111] The above-mentioned change points can serve as the key basis for distinguishing between stable segments and non-stationary segments. The stable segment usually appears as an area with a small slope and gentle changes, while the non-stationary segment appears as an area with a large slope and a trend of rapid changes.

[0112] According to the speed range corresponding to the stable segment, the geometric offset of the magnetic field center in the corresponding range is extracted from the speed-offset curve, and the average of these offsets is calculated and recorded as the reference offset.

[0113] It should be noted that if there are multiple stationary segments, each stationary segment corresponds to a reference offset. To ensure accuracy, when calculating the differential offset in a non-stationary segment, the reference offset corresponding to the stationary segment closest to the speed point of the non-stationary segment should be selected for comparison.

[0114] It should be understood that in the stable section, the motor is in a steady-state operation state, and the vibration has little effect on the center position of the magnetic field. Therefore, the average of the offsets in these intervals can be used as the reference offset.

[0115] According to the speed points existing in the non-stationary segment, the geometric offset of the magnetic field center corresponding to each speed point is extracted from the speed-offset curve, and the difference offset is obtained by subtracting the geometric offset from the reference offset.

[0116] It's important to note that factors such as electromagnetic interference, temperature fluctuations, and vibration can affect the magnetic field center during motor operation. However, in the dynamic testing environment of this invention, by optimizing experimental conditions, the potential impact of external factors such as electromagnetic interference and temperature changes on the test results has been minimized. Based on this, the differential offset is used as a key indicator to quantify the specific impact of vibration on the magnetic field center offset, highlighting the unique contribution of vibration to the magnetic field center offset.

[0117] The differential offset corresponding to each speed point and the motor vibration data are cross-correlated to obtain the cross-correlation coefficient, which is then compared with the preset effective cross-correlation coefficient. If the cross-correlation coefficient reaches the correlation coefficient threshold, it is predicted that there is a significant correlation between the geometric offset of the magnetic field center position and the vibration.

[0118] The cross-correlation calculation described above can quantify the degree of correlation between the differential offset and the motor vibration. Specifically, this method is used to evaluate whether the differential offset also exhibits a corresponding change trend when the motor vibration characteristics change.

[0119] In an improved implementation of the above solution, a Fast Fourier Transform (FFT) can also be used to compare the differential offset and the motor vibration data spectra to identify common frequency components. If certain frequency components in the two spectra overlap significantly, it indicates that the vibration has a significant impact on the magnetic field distribution.

[0120] In a further innovative implementation of the above scheme, the vibration correction of the magnetic field center offset under dynamic detection is carried out according to the following process: a regression equation is established for the differential offset and motor vibration data corresponding to each speed point in the non-stationary section, and the regression coefficient is fitted and solved to represent the offset change caused by unit vibration.

[0121] It's important to understand that the regression equation describes the mathematical relationship between motor vibration data, such as acceleration and amplitude, and the offset. This relationship can be linear or nonlinear, depending on the actual physical mechanism. Regression analysis quantifies the specific impact of vibration on the magnetic field center offset. The regression coefficient clearly defines the specific impact of a unit vibration change on the magnetic field center position.

[0122] The regression coefficient is used to calculate the offset component caused by vibration based on the motor vibration data corresponding to each speed point in the non-stationary segment.

[0123] The correction offset is obtained by deducting the offset component caused by vibration from the geometric offset of the magnetic field center corresponding to each speed point in the non-stationary segment.

[0124] It's important to note that the corrected offset obtained through regression analysis is based on subtracting the vibration-induced offset from the geometric offset of the magnetic field center. To verify the effectiveness of the correction, the residuals between the corrected and original offsets can be evaluated and their statistical properties, such as standard deviation, analyzed. A significant reduction in the residuals indicates that the vibration correction measures were effective.

[0125] S3. Detection fusion: The magnetic field center offset after static and dynamic detection correction is integrated and weighted fusion is used to generate the final offset.

[0126] Specifically, the magnetic field center offset after static and dynamic detection correction is integrated and weighted fusion is used to generate the final offset, which includes the following: extracting the maximum offset and the average offset from the magnetic field center geometric offset corresponding to each speed point after dynamic detection correction, and calculating the standard deviation of the magnetic field center geometric offset corresponding to all speed points, which is recorded as the dynamic offset standard deviation.

[0127] Understandably, the maximum offset represents the maximum deviation of the magnetic field center across the entire speed range. It reflects the maximum possible deviation during motor operation. The average offset is the mean of the magnetic field center offset at all speed points, providing an overall picture of the offset across the speed range. The dynamic offset standard deviation measures the dispersion of the magnetic field center offset at each speed point. A smaller standard deviation indicates a relatively concentrated and stable offset, while a larger standard deviation may indicate significant fluctuation or inconsistency.

[0128] Similarly, according to the method of obtaining the static magnetic field center offset, the dynamic offset standard deviation is used to weight the maximum offset and the average offset to obtain the magnetic field center offset after dynamic detection correction.

[0129] The final offset is obtained by taking a weighted average of the static magnetic field center offset and the magnetic field center offset after dynamic detection correction.

[0130] The weighting of these two parameters can be configured based on the specific application environment. Given the higher likelihood of magnetic field center shift in dynamic operating environments, the offset after dynamic detection correction is typically assigned a higher weight, while the static offset is assigned a lower weight. This approach enhances the precise capture of magnetic field center position changes under actual operating conditions, improving the effectiveness and accuracy of the calculation results.

[0131] The above embodiments may be implemented in whole or in part through software, hardware, firmware or any other combination. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0132] Those skilled in the art will appreciate that the algorithmic steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0133] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0134] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting the magnetic field center of a motor, characterized in that: The following steps are involved: S1. Static test: S11. Secure the motor and expose the stator core and rotor poles. Align the magnetic imaging terminal with the motor pole axis. S12. Using a magnetic imaging terminal to collect static magneto-optical images, and extracting the magnetic field center position and at least three reference stator positions; S13. Analyze the static magnetic field center offset by comparing the geometric offset of the magnetic field center with each reference stator position; S2. Dynamic detection: S21. Start the motor and dynamically increase the speed in steps to the target speed, synchronously capturing the magneto-optical image sequence and motor vibration data sequence at each speed point; S22. Ghosting recognition is performed on the magneto-optical image sequence at each speed point, and the magneto-optical image and synchronous motor vibration data meeting the standards are screened accordingly; S23. Extract the magnetic field center position from the standard magneto-optical image corresponding to each speed point and compare it with the same reference stator position to calculate the dynamic geometric offset, and perform correlation analysis with the synchronous motor vibration data. Based on this, perform vibration correction on the magnetic field center offset under dynamic detection; S3. Detection fusion: The magnetic field center offset after static and dynamic detection correction is integrated and weighted fusion is used to generate the final offset.

2. The method for detecting the magnetic field center of a motor according to claim 1, wherein: The extraction process of the at least three reference stator positions is as follows: Delimiting a closed boundary region in a static magneto-optical image according to actual positions of a plurality of stators of the motor; The geometric center point is calculated within the closed boundary area, and starting from the center point, multiple directional rays are diverged radially toward the boundary line at uniform intervals to form multiple directional rays. The stator position corresponding to the intersection point of each directional ray and the boundary line is regarded as the reference stator position.

3. The method for detecting the magnetic field center of a motor according to claim 1, wherein: The analysis of the static magnetic field center offset is as follows: Compare the geometric offsets of the magnetic field center with those of each reference stator position, select the maximum offset and the average offset, and calculate the standard deviation of the geometric offsets of all reference stator positions, which is recorded as the static offset standard deviation; Normalize the static offset standard deviation; The maximum offset and the average offset are weighted using the normalized static offset standard deviation. The static magnetic field center offset is obtained by combining the maximum offset, the average offset and their corresponding weights for weighted calculation.

4. The method for detecting the magnetic field center of a motor according to claim 1, wherein: The starting of the motor and dynamically increasing the speed to the target speed in steps of a preset speed includes the following: Set a speed range based on the motor's specifications and divide the entire speed range into three intervals: low speed, medium speed, and high speed. The set ratio of the motor's rated speed is used as the preset speed step; Different multiple factors are used to divide discrete speed points according to different speed ranges, as follows: a) Divide the low-speed zone into several discrete speed points according to the preset speed step size; b) Divide the medium speed zone into a number of discrete speed points according to multiples of the preset speed step length; c) Divide the high-speed area into several discrete speed points according to the increasing multiples of the preset speed step length; Arrange all the discrete speed points obtained by division in ascending order to form the final speed sequence; Start the motor and accelerate to the corresponding speed value according to the discrete speed points arranged above.

5. The method for detecting the magnetic field center of a motor according to claim 1, wherein: The synchronous capture of the magneto-optical image sequence and the motor vibration data sequence at each speed point is implemented as follows: Set a steady-state detection time window at each speed point; When the actual speed of the motor after starting enters the allowable error range of each speed point, the motor operating indicators of adjacent times are collected within the set steady-state detection time window and compared with the set critical difference. If the differences of all motor operating indicators of multiple consecutive adjacent times are less than or equal to the critical difference, the current speed is locked for a fixed period of time; During the locking period of each speed point, magneto-optical image acquisition and vibration detection are performed synchronously on the motor to obtain multiple frames of magneto-optical images and multiple motor vibration data corresponding to each speed point, and the magneto-optical image sequence and motor vibration data sequence are arranged in chronological order.

6. A method for detecting the magnetic field center of a motor according to claim 5, characterized in that: The ghost recognition of the magneto-optical image sequence at each speed point is performed as follows: The magneto-optical image sequence at each rotation speed point is subjected to denoising and contrast enhancement processing; Edge extraction is performed on the preprocessed magneto-optical image sequence. If multiple edges are extracted from a magneto-optical image, ghosting is identified and the image with multiple edges is marked as a ghosting image.

7. A method for detecting the magnetic field center of a motor according to claim 6, characterized in that: The contents of the screening qualified magneto-optical images and synchronous motor vibration data are as follows: Remove ghost images from the magneto-optical image sequence at each speed point. If any remaining image exists, it is regarded as a qualified magneto-optical image, and the timestamp of the qualified magneto-optical image is recorded. Then, based on the timestamp, the motor vibration data corresponding to the timestamp is extracted from the collected multiple motor vibration data as the synchronous motor vibration data; If there are no remaining images, the magneto-optical image sequence is processed through multi-frame synthesis to generate a qualified magneto-optical image. At this time, the motor vibration data corresponding to the middle timestamp of multiple motor vibration data at the corresponding speed point is selected as the synchronous motor vibration data.

8. The method for detecting the magnetic field center of a motor according to claim 1, wherein: The process of correlation analysis with synchronous motor vibration data is as follows: For the geometric offset of the magnetic field center at each speed point, a speed-offset curve is drawn in a coordinate system with the speed value corresponding to the speed point as the horizontal axis and the geometric offset as the vertical axis; Similarly, the speed-vibration curve is drawn based on the synchronous motor vibration data corresponding to each speed point; Calculate the slopes of the data points on the speed-offset curve and the speed-vibration curve respectively, and mark the points where the absolute value of the slope reaches the configured threshold as change points; The speed-vibration curve is divided into a stable segment and a non-stationary segment based on the change points marked on the speed-vibration curve; According to the speed range corresponding to the stable segment, the geometric offset of the magnetic field center in the corresponding range is extracted from the speed-offset curve, and the average of these offsets is calculated and recorded as the reference offset; According to the speed points in the non-stationary section, the geometric offset of the magnetic field center corresponding to each speed point is extracted from the speed-offset curve, and the difference offset is obtained by subtracting the geometric offset from the reference offset; The differential offset corresponding to each speed point and the motor vibration data are cross-correlated to obtain the cross-correlation coefficient, which is then compared with the preset effective cross-correlation coefficient. If the cross-correlation coefficient reaches the correlation coefficient threshold, it is predicted that there is a significant correlation between the geometric offset of the magnetic field center position and the vibration.

9. The method for detecting the magnetic field center of a motor according to claim 8, wherein: The vibration correction of the magnetic field center offset under dynamic detection is performed as follows: A regression equation is established for the difference offset and motor vibration data corresponding to each speed point in the non-stationary section, and the regression coefficient is fitted to represent the offset change caused by unit vibration; The regression coefficient is used to calculate the offset component caused by vibration based on the motor vibration data corresponding to each speed point in the non-stationary segment; The correction offset is obtained by deducting the offset component caused by vibration from the geometric offset of the magnetic field center corresponding to each speed point in the non-stationary segment.

10. The method for detecting the magnetic field center of a motor according to claim 3, wherein: The magnetic field center offset after the integrated static and dynamic detection correction is generated by weighted fusion to generate the final offset, which includes the following: Extract the maximum offset and average offset from the geometric offset of the magnetic field center corresponding to each speed point after dynamic detection correction, and calculate the standard deviation of the geometric offset of the magnetic field center corresponding to all speed points, which is recorded as the dynamic offset standard deviation; Similarly, according to the method of obtaining the static magnetic field center offset, the dynamic offset standard deviation is used to weight the maximum offset and the average offset to obtain the magnetic field center offset after dynamic detection correction; The final offset is obtained by taking a weighted average of the static magnetic field center offset and the magnetic field center offset after dynamic detection correction.

Images