CN117195752A

CN117195752A - Modeling method of axial split-phase hybrid excitation type magnetic levitation motor

Info

Publication number: CN117195752A
Application number: CN202311470544.XA
Authority: CN
Inventors: 袁轶彦; 尚栋; 王春彦
Original assignee: Suzhou Baobang Electric Co ltd
Current assignee: Suzhou Baobang Electric Co ltd
Priority date: 2023-11-07
Filing date: 2023-11-07
Publication date: 2023-12-08
Anticipated expiration: 2043-11-07
Also published as: CN117195752B

Abstract

The invention relates to the technical field of three-dimensional data modeling, in particular to a modeling method of an axial split-phase hybrid excitation type magnetic levitation motor. The method comprises the following steps: acquiring a magnetic suspension motor image set; extracting shell characteristic points from the magnetic suspension motor image set to generate motor shell characteristic point data; three-dimensional reconstruction is carried out on the characteristic point data of the motor shell, and a three-dimensional model of the motor shell is generated; acquiring data of a magnetic suspension motor; performing motor mixed excitation analysis on the magnetic suspension motor data to generate output torque data of the suspension motor and rotating speed data of the suspension motor; carrying out hysteresis deviation value analysis on the output torque data of the suspension motor and the rotating speed data of the suspension motor to generate hysteresis deviation data; according to the invention, by utilizing image processing, data analysis, environmental interference compensation and axial split-phase control, the error influence of hysteresis effect on the modeling of the suspension motor is solved, so that the modeling accuracy is improved.

Description

Modeling method of axial split-phase hybrid excitation type magnetic levitation motor

Technical Field

The invention relates to the technical field of three-dimensional data modeling, in particular to a modeling method of an axial split-phase hybrid excitation type magnetic levitation motor.

Background

The axial split-phase mixed excitation type magnetic suspension motor (hereinafter referred to as mixed excitation type motor) is an advanced motor technology, combines the axial split-phase and magnetic suspension technologies, and has the advantages of high efficiency, low noise, low maintenance and the like. The magnetic suspension technology starts to rise in the middle of the 20 th century, and utilizes a magnetic field to suspend a rotating part in a non-contact state, so that mechanical abrasion and friction are eliminated, the efficiency and service life of a motor are improved, and axial split phase is a configuration method of a motor winding, current is divided into a plurality of phases, and the motor is more stable and efficient. The introduction of the technology improves the performance of the motor, reduces the energy loss, and the hybrid excitation motor is an innovation combining the magnetic suspension technology and the axial phase separation technology, and introduces the axial phase separation on the basis of the magnetic suspension motor, so that the motor is more efficient, stable and controllable, however, the magnetic field characteristic and the hysteresis effect cannot be well evaluated in the modeling process of the current hybrid excitation, so that the data detection of a rotor is not accurate enough in the modeling of the hybrid excitation, and the modeling accuracy is lower.

Disclosure of Invention

Based on this, it is necessary to provide a modeling method for an axial split-phase hybrid excitation type magnetic levitation motor, so as to solve at least one of the above technical problems.

In order to achieve the above purpose, a modeling method of an axial split-phase hybrid excitation type magnetic levitation motor comprises the following steps:

step S1: acquiring a magnetic suspension motor image set; extracting shell characteristic points from the magnetic suspension motor image set to generate motor shell characteristic point data; three-dimensional reconstruction is carried out on the characteristic point data of the motor shell, and a three-dimensional model of the motor shell is generated;

step S2: acquiring data of a magnetic suspension motor; performing motor mixed excitation analysis on the magnetic suspension motor data to generate output torque data of the suspension motor and rotating speed data of the suspension motor; carrying out hysteresis deviation value analysis on the output torque data of the suspension motor and the rotating speed data of the suspension motor to generate hysteresis deviation data; constructing a hybrid excitation model through hysteresis deviation data, output torque data of the suspension motor and rotating speed data of the suspension motor, so as to generate a hybrid excitation mathematical model;

step S3: the suspension motor outputs torque data and suspension motor rotating speed data to carry out noise detection on the suspension motor, and motor operation noise data is generated; performing interference compensation analysis on the sensor according to the motor operation noise data to generate an operation interference compensation coefficient; performing axial phase separation on the suspension motor according to the operation interference compensation coefficient, the suspension motor output torque data and the suspension motor rotating speed data to obtain an axial phase separation mathematical model;

Step S4: performing model coupling on the axial split-phase mathematical model, the mixed excitation mathematical model and the motor shell three-dimensional model to generate a three-dimensional coupling model; carrying out motor component interconnection analysis on the three-dimensional coupling model to generate motor component interconnection data;

step S5: model training is carried out on the motor component interconnection data, and a motor performance prediction model is generated; and performing control optimization and visualization on the three-dimensional coupling model by using the motor performance prediction model so as to generate an axial split-phase hybrid excitation type magnetic suspension motor modeling scheme.

The invention collects the image data set of the magnetic levitation motor and can be acquired by a camera, a scanner or other imaging equipment. The image set should cover various angles and portions of the motor to ensure integrity, and feature points of the housing are extracted from the image set of the maglev motor using an image processing algorithm. These feature points may include edges, corners, etc. of the motor profile. The accuracy of feature point extraction directly influences the effect of subsequent three-dimensional reconstruction, and shell feature points extracted from images are integrated into a data set. The data may include coordinates, color information, etc. of each feature point, and on the basis of the three-dimensional reconstruction, a complete three-dimensional model of the motor housing is generated. This model may be used for further analysis, design, presentation, etc. applications. The generated three-dimensional model can more comprehensively and intuitively present the appearance and the structure of the magnetic levitation motor, and can realize the three-dimensional modeling of the magnetic levitation motor, thereby providing more comprehensive and visual information. And analyzing an excitation system of the magnetic suspension motor. Hybrid excitation may include different control strategies and excitation patterns, such as a combination of current and magnetic levitation force. From this analysis, output torque data and rotational speed data of the motor can be obtained, and hysteresis refers to a nonlinear response of the magnetic material in the motor field, usually accompanied by hysteresis losses. The purpose of the hysteresis bias analysis is to understand the effect of hysteresis on motor performance. The step can generate hysteresis deviation data, describe the nonlinear behavior of the magnetic field and can realize the deep understanding of the behavior and the performance of the magnetic levitation motor. The construction of the hybrid excitation model allows engineers and researchers to better control and optimize the operation of the motor. Knowing the effect of hysteresis bias on motor performance helps to improve motor design and control strategies. Such models may also be used to simulate, predict motor behavior, and optimize motor performance in practical applications. By analyzing the torque data and the rotational speed data, noise in the operation of the motor can be detected. This may be due to mechanical vibration, electronic noise or other sources of interference. Generating operational noise data that can be used to analyze sensor disturbances can help to understand the operational quality and stability of the motor. This includes detecting errors or disturbances that the sensor may introduce, such as noise, drift or calibration problems of the sensor. Through the analysis, an interference compensation coefficient can be generated for correcting sensor data, measuring accuracy is improved, and an axial split-phase mathematical model of the suspension motor can be generated by using the running interference compensation coefficient, suspension motor output torque data and suspension motor rotating speed data. This model describes the behavior of the motor under different operating conditions, taking into account the correction of noise and sensor disturbances. This helps to more accurately predict the performance and behavior of the motor. Through more comprehensive motor modeling and component interconnection analysis, the performance characteristics of the motor can be better understood, and optimized to improve efficiency, reduce energy consumption or meet specific performance requirements, and analysis of interconnection data between motor components can be used to detect potential faults or problems. This facilitates early warning and maintenance, and by in-depth analysis of interactions between various components of the motor, valuable insights can be obtained as to how to improve the design or manufacture of the motor, facilitate a more comprehensive understanding of the motor system, optimize performance, improve reliability, and provide important information as to how the various components interact. The motor component interconnection data is used for training a machine learning or mathematical model, the model can predict the performance of the motor, the generated model can be used for rapidly predicting the performance of the motor without relying on complex simulation or experiments, and the model can be used for evaluating the influence of different design choices on the performance of the motor, so that the design iteration is performed more rapidly, the requirements on actual motor manufacturing and testing are reduced, and the cost is reduced. Therefore, the invention solves the error influence of hysteresis effect on the modeling of the suspension motor by utilizing image processing, data analysis, environmental interference compensation and axial split-phase control, thereby improving the modeling accuracy.

The method has the beneficial effects that the three-dimensional model of the motor shell can be generated by acquiring the shell image set of the motor, extracting the shell characteristic points and carrying out three-dimensional reconstruction. This helps to better understand the structural and geometric characteristics of the motor by analyzing the motor data to generate a hybrid excitation mathematical model. This model can be used to predict the output torque and speed of the motor, as well as the effect of hysteresis bias. This is important for optimization and control of motor performance, and by detecting motor output torque data and rotational speed data, and analyzing motor operating noise data, an operating disturbance compensation coefficient can be generated. The method is beneficial to reducing the influence of noise on the motor performance, improving the accuracy and reliability of the system, and the axial phase separation model comprises the step of axially separating phases of the motor to obtain an axial phase separation mathematical model. The method is helpful for better understanding the axial characteristics and performances of the motor, so that the control and the optimization are better performed, and the axial split-phase model, the hybrid excitation model and the motor shell three-dimensional model are coupled to generate a comprehensive three-dimensional coupling model. The model can be used for analyzing interconnection relations among different components, further optimizing motor performance, and generating a motor performance prediction model by performing model training on motor component interconnection data. This model can be used to predict motor performance under different operating conditions, helping control optimization. In addition, the visualization may help engineers better understand the operation of the motor. Therefore, the invention solves the error influence of hysteresis effect on the modeling of the suspension motor by utilizing image processing, data analysis, environmental interference compensation and axial split-phase control, thereby improving the modeling accuracy.

Drawings

FIG. 1 is a schematic flow chart of the modeling method of an axial split-phase hybrid excitation type magnetic levitation motor;

FIG. 2 is a flowchart illustrating the detailed implementation of step S2 in FIG. 1;

FIG. 3 is a flowchart illustrating the detailed implementation of step S24 in FIG. 2;

FIG. 4 is a flowchart illustrating the detailed implementation of step S3 in FIG. 1;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

To achieve the above objective, please refer to fig. 1 to 4, a modeling method of an axial split phase hybrid excitation magnetic levitation motor, the method includes the following steps:

In the embodiment of the present invention, as described with reference to fig. 1, the modeling method of the axial split-phase hybrid excitation type magnetic levitation motor of the present invention includes the following steps:

in the embodiment of the invention, by collecting a group of images of the magnetic levitation motor, the images may be taken from different angles and distances to ensure that the appearance of the motor housing is captured comprehensively, and using computer vision technology, such as a feature point detection algorithm (e.g. SIFT, SURF, ORB, etc.), feature points are extracted from each image, and the feature points should be salient points on the motor housing, such as edges, corner points, etc. These feature points are used in subsequent processing, and for each image, the coordinates and descriptors of the extracted feature points are saved into a data structure to create a feature point dataset. The descriptor is a numerical representation of the appearance of the feature points, and the extracted feature point data is fused from different images using a three-dimensional reconstruction technique to generate a three-dimensional model of the motor housing. This can be accomplished using structured light scanning, stereo vision, multi-view geometry, etc., and the feature point data can be converted into a three-dimensional model using three-dimensional point cloud reconstruction or surface reconstruction methods. Integrating all three-dimensional data points to create a three-dimensional model of the complete motor housing, which can be further processed, such as texture mapping, multi-resolution model generation, color processing, etc., to improve the visualization effect.

in the embodiment of the invention, the data related to the magnetic suspension motor is acquired. Such data may include motor operating data such as input current, voltage, temperature, vibration, etc. of the motor. The data can be acquired through a sensor, a monitoring system or other data acquisition methods, and the acquired motor data is subjected to mixed excitation analysis. Hybrid excitation is typically a motor control strategy that involves controlling the motor using different excitation patterns simultaneously to achieve specific performance objectives. This may include using magnetic levitation and conventional electromagnetic windings to manipulate the motor, and during analysis, generating output torque data and rotational speed data for the levitated motor. These data are used to understand the performance and behavior of the motor, and the hysteresis bias is the hysteresis effect of the magnetic field variations in the magnetic material. In this step, the output torque and rotational speed data of the motor is analyzed to identify and quantify effects associated with hysteresis bias. The generation of hysteresis deviation data is helpful for better understanding the motor behavior, and a mixed excitation mathematical model is constructed by utilizing the hysteresis deviation data generated before and the output torque and rotation speed data of the levitation motor. This model may be used to predict motor performance, optimize motor control strategies, or perform other relevant analyses.

in the embodiment of the invention, noise detection is carried out on the output torque data and the rotating speed data of the suspension motor. This aims to identify and quantify noise or interference associated with motor operation. Noise detection may involve the use of signal processing techniques, such as filtering or spectral analysis, to distinguish between signal and noise components, and interference compensation analysis based on previously generated motor operation noise data. This may involve determining the disturbance characteristics of the individual sensors and their effect on the motor data. And generating operation interference compensation coefficients according to the analysis result, wherein the coefficients are used for reducing or eliminating the influence of sensor interference on data, and performing axial phase separation analysis by using the operation interference compensation coefficients generated before and output torque data and rotation speed data of the suspension motor. This aims at determining the axial split-phase behaviour of the motor, i.e. how the output of the motor varies with the axial position. Based on the analysis results, an axial split-phase mathematical model is constructed, which can help understand and predict the performance and behavior of the motor.

in embodiments of the present invention, an overall three-dimensional coupled model is generated by coupling together different models previously created. This means that the axial split-phase mathematical model, the hybrid excitation mathematical model and the motor housing three-dimensional model are integrated into one common frame. This may involve connecting the inputs and outputs of the models together so that they can interact and co-operate, and once the three-dimensional coupling model is generated, the next step is to perform motor part interconnection analysis on the model. This includes taking into account interactions and interactions between components inside the motor. This may include electromagnetic components of the motor, mechanical components, control systems, etc. From this analysis, interconnection data between motor components can be generated to better understand the behavior of the overall motor system.

In the embodiment of the invention, a motor performance prediction model is trained by using motor component interconnection data. This may be a machine learning model, such as a neural network, or a mathematical model based on physical principles. The goal of the training model is to be able to accurately predict motor performance, such as output torque, speed, efficiency, etc., based on input parameters and operating conditions, once the motor performance prediction model is trained, the model can be used for control optimization. This means that the control strategy of the motor is optimized to meet specific performance requirements based on the prediction results of the model. This may include controlling parameters of the motor such as current, voltage, excitation pattern, etc. to obtain optimal performance, visualization is an important tool that may help understand the output of the model and optimize the results. Through visualization, the performance, the running condition and the control strategy of the motor can be intuitively displayed. This may include generating charts, animations, or other visualization tools to better understand the motor system by engineers and decision makers.

Preferably, step S1 comprises the steps of:

step S11: acquiring a multi-view magnetic levitation motor image set;

step S12: performing image preprocessing on the magnetic levitation motor image set, wherein the image preprocessing comprises image transformation, image enhancement and image texture analysis, so as to obtain a standard magnetic levitation motor image set;

Step S13: performing image segmentation on a standard magnetic levitation motor image set to generate a magnetic levitation motor segmented image set; extracting motor shell edge feature points from the magnetic suspension motor segmentation image set to generate motor shell feature point data, wherein the motor shell feature point data comprise motor shell edge data and motor shell corner data;

step S14: performing feature point matching on the magnetic suspension motor according to the motor housing edge data and the motor housing corner data to generate motor housing feature matching data; performing feature point three-dimensional space mapping on the motor shell feature matching data by a camera calibration method to generate sparse three-dimensional point cloud data;

step S15: dense three-dimensional reconstruction is carried out on the sparse three-dimensional point cloud data, and dense three-dimensional point cloud data are generated; and modeling the motor shell by utilizing three-dimensional modeling software to perform motor shell modeling on the dense three-dimensional point cloud data to generate a motor shell three-dimensional model.

According to the invention, through acquiring the multi-view image set, the multi-angle motor image can be acquired, so that more comprehensive information and view angles can be provided, the appearance and structure of the motor can be better understood, and the image preprocessing is conducive to reducing noise and enhancing image characteristics, so that the quality of subsequent processing is improved. The processing can enable the standard image set to be easier to analyze so as to facilitate modeling of the motor shell, and key features such as shell edges and corner points of the motor can be determined by dividing the standard image set and extracting the feature points. This helps to capture the external shape and structural information of the motor, and by matching the feature points of different perspectives, a three-dimensional spatial model of the motor housing can be created. This facilitates integration of information from different perspectives into a common three-dimensional coordinate system for subsequent three-dimensional modeling, and converting sparse three-dimensional point cloud data into dense three-dimensional point clouds facilitates more accurate capturing of details of the motor housing. These data can be used for further three-dimensional modeling, with the generated three-dimensional model of the motor housing being the primary target of the process. This model can be used to analyze the structure of the motor, design improvements, and integration with other systems.

In the embodiment of the invention, through the image pickup equipment, such as a camera or a video camera, to acquire a plurality of motor images with different angles, a mechanical device or a manual operation camera can be used for acquiring multi-view images, so that all aspects of the motor are covered, the quality and the resolution of image acquisition are ensured, and a clear image is acquired. Image preprocessing may be accomplished using image processing software or programming languages (e.g., the OpenCV library of Python), image transformation may include rotating, cropping, or flipping the images to ensure that they are aligned under the same coordinate system, image enhancement may include brightness, contrast, sharpening, etc. adjustments to improve image quality, and image texture analysis may include texture feature extraction, such as Gabor filters, etc., to identify texture information in the images. Image segmentation may use image processing algorithms (e.g., thresholding, edge detection, watershed segmentation, etc.) to segment the motor image into different regions, such as motor housing and other parts, and feature point extraction may use feature detection algorithms (e.g., harris corner detection, SIFT, SURF, etc.) to extract edges and corners of the motor housing, generating motor housing feature point data, including preserving the edge data and the corner data. The feature point matching can use feature descriptors (such as SIFT descriptors and ORB descriptors) to match the feature points of the motor shell at different visual angles, the camera calibration method can use a camera calibration plate and a calibration algorithm (such as a Zhang's calibration method) to obtain internal parameters and external parameters of a camera, the feature points of the motor shell are mapped to a three-dimensional space through feature point matching and camera calibration, and sparse three-dimensional point cloud data are generated. The sparse three-dimensional point cloud data can be converted into dense three-dimensional point cloud data through a three-dimensional reconstruction algorithm (such as structured light scanning, stereoscopic vision, point cloud fusion and the like), and modeling of the motor shell can be performed on the dense three-dimensional point cloud data by utilizing three-dimensional modeling software (such as Blender, autoCAD, solidWorks and the like) to generate a three-dimensional model of the motor shell, and textures, colors and other details can be added in the modeling process to obtain a realistic appearance.

Preferably, step S2 comprises the steps of:

step S21: acquiring data of a magnetic suspension motor;

step S22: carrying out data preprocessing on the magnetic levitation motor data, wherein the data preprocessing comprises data cleaning, data protocol, data conversion and data standardization, so as to obtain standard magnetic levitation motor data;

step S23: screening the motion structure data of the standard magnetic levitation motor data to obtain motor rotor motion data and motor stator data; performing motor magnetic field distribution simulation on motor rotor motion data and motor stator data through a finite element analysis tool to generate motor simulation magnetic field distribution data;

step S24: performing simulation operation on the magnetic suspension motor according to the motor simulation magnetic field distribution data, and acquiring power operation data of the magnetic suspension motor by utilizing a power sensor; carrying out mixed excitation on the magnetic levitation motor based on the electric power operation data of the magnetic levitation motor and the motor simulation magnetic field distribution data and collecting data, so as to obtain output torque data of the magnetic levitation motor and rotating speed data of the magnetic levitation motor;

step S25: performing magnetic field dependency detection on the motor simulation magnetic field distribution data by utilizing an electromagnetic dependency energy efficiency analysis formula through the output torque data of the levitation motor and the rotating speed data of the levitation motor to generate magnetic field dependency data; comparing the magnetic field dependency data with a preset historical magnetic field dependency threshold, and marking the corresponding magnetic field dependency data as stable magnetic field data when the magnetic field dependency data is larger than or equal to the historical magnetic field dependency threshold; when the magnetic field dependency data is smaller than the historical magnetic field dependency threshold value, marking the corresponding magnetic field dependency data as unstable magnetic field data;

Step S26: establishing a hysteresis effect model through unstable magnetic field data to generate a hysteresis effect mathematical model; introducing the stable magnetic field data into a hysteresis effect mathematical model to analyze hysteresis deviation values and generate hysteresis deviation data; and constructing a hybrid excitation model through hysteresis deviation data, output torque data of the levitation motor and rotating speed data of the levitation motor, so as to generate a hybrid excitation mathematical model.

The invention collects data from the magnetic levitation motor, wherein the data may comprise information such as operation data, sensor data, current, voltage, rotating speed and the like, the data can be acquired through a sensor, a data recording device or other data acquisition methods, the data preprocessing is used for ensuring the quality and consistency of the data, the data cleaning comprises the processing of missing values, abnormal values and noise so as to ensure the accuracy of the data, the data specification can reduce the dimension of the data and redundant information, the data conversion may comprise the mathematical transformation or transformation of the data into an applicable data format, and the data standardization can ensure the consistency of the data under different scales so as to support the subsequent analysis and modeling. The finite element analysis tool is used for processing the standard magnetic suspension motor data, extracting the motor rotor and stator data, and simulating the magnetic field distribution inside the motor, which is very important for motor performance analysis and design optimization. The simulation operation is carried out on the motor according to the simulated magnetic field distribution data of the motor, the motor performance can be evaluated, the electric power operation data of the motor, such as output torque and rotation speed, can be obtained by utilizing equipment such as an electric force sensor, and the like, and the data acquisition can be carried out under different working conditions in a mixed excitation mode so as to obtain more information. The electromagnetic dependence energy efficiency analysis formula is used, the motor output torque and the rotating speed data are combined to detect the magnetic field dependence of the motor, and the data are marked as stable or unstable magnetic field data by comparing with historical data, so that the performance and the stability of the motor can be predicted. Based on the unstable magnetic field data, a hysteresis effect mathematical model is constructed to know the influence of hysteresis effect in the motor on performance, and a hybrid excitation mathematical model is constructed by using hysteresis deviation data, output torque data and rotation speed data to better understand the working principle and performance of the motor.

As an example of the present invention, referring to fig. 2, the step S2 in this example includes:

step S21: acquiring data of a magnetic suspension motor;

in the embodiment of the invention, the performance of the motor is monitored by installing various sensors. These sensors may include current sensors, voltage sensors, rotational speed sensors, temperature sensors, displacement sensors, and the like. These sensors may be directly connected to critical parts of the motor to acquire data, and a data acquisition system is used to capture and record the sensor data. These systems may be hardware specifically designed for motor monitoring, or may be computer software that communicates with the sensors and records data via an interface. Typically, the data acquisition system has storage, real-time monitoring and data processing functions, and appropriate communication protocols are used to connect the sensors to the data acquisition system. This may involve different communication modes of analog signals, digital signals, ethernet, CAN bus, etc., depending on the type of sensor and data acquisition system, determining the frequency of data acquisition, i.e. the time interval of data recording. Depending on the parameters to be monitored and the requirements of the application. Some parameters may require high frequency acquisition while others may require only low frequency acquisition, with the acquired data stored in a suitable storage medium, such as a hard disk drive, cloud storage, or database. The data should be stored in a retrievable and manageable manner for future analysis and reference to obtain the magnetic levitation motor data.

in the embodiment of the invention, the data preprocessing is carried out on the data of the magnetic levitation motor, which comprises the processing of missing values, abnormal values and noise. For missing values, missing data points may be processed by interpolation or deletion. Outliers and noise may be identified and processed by statistical or rule-based methods, such as smoothing techniques or filtering methods, and if the amount of data is large, data may be considered for reduction of complexity and storage requirements of the data. Data reduction may be implemented by sampling, aggregation, or dimension reduction techniques, such as Principal Component Analysis (PCA) or Linear Discriminant Analysis (LDA), and the like, sometimes requiring conversion of the raw data to meet specific analysis requirements or algorithm preconditions. Common data transformations include logarithmic transformation, square root transformation, normalization, and the like. These transformations help reduce the variance between the data, making the data more desirable for the analytical model, and the normalization is to scale the data to a specific range, helping to eliminate unit of measure differences between different features, to ensure that the data can be effectively compared and processed in a statistical analysis or machine learning model. Common normalization methods include Z-score normalization and min-max normalization, among others.

in the embodiment of the invention, the data related to the motor rotor motion and the motor stator are firstly needed to be screened out from the standard magnetic levitation motor data. This may include information about location, shape, magnetic properties, material properties, etc. It is important to ensure accuracy and integrity of the data and to select appropriate finite element analysis tools, such as ANSYS, COMSOL Multiphysics, FEMM (Finite Element Method Magnetics), etc., for motor field distribution simulation. The selected tools are installed and configured to meet simulation requirements, and a model of the motor is built using a finite element analysis tool. This includes defining the geometry, material properties, boundary conditions and excitation conditions of the motor. In general, a three-dimensional geometric model needs to be built in order to accurately simulate the magnetic field distribution inside the motor, and the motor model is gridded to be decomposed into small finite element units. The degree of refinement of the grid can affect the accuracy of the simulation and the computational complexity. In general, moderate meshing is required to balance accuracy and computational efficiency, with appropriate physical properties such as permeability, conductivity, magnetic properties, etc. being assigned to each part in the model. These properties affect the results of the simulation and therefore need to be set accurately according to the actual situation, and the excitation conditions are defined according to the operation of the motor. This may include current excitation, magnetic field excitation, or other physical excitation. Ensuring that the excitation conditions are consistent with motor data, and using a finite element analysis tool to run the magnetic field distribution simulation of the motor. This will calculate the magnetic field distribution inside the motor, including the magnetic induction and magnetic field profile.

in the embodiment of the invention, the operation of the magnetic suspension motor is simulated by using the motor simulation magnetic field distribution data obtained before in a simulation environment. This may include simulating the start, run and stop processes of the motor. In the simulation, various operating conditions of the motor, including current, voltage, load variation, etc., are simulated, and power sensors, typically a torque sensor and a rotation speed sensor, are installed to monitor power operation data of the magnetic levitation motor. These sensors are typically coupled to bearings or other associated parts of the motor to measure torque and rotational speed, and connect the electrical sensors to a data acquisition system that can record the output data of the sensors. The data acquisition system can be ensured to acquire data with high precision and high frequency so as to capture details in the operation process of the motor, the motor is started in simulation, and the data acquisition system is started simultaneously so as to record the power operation data of the motor. These data include torque, rotational speed, current, voltage, etc. The acquisition frequency of the data is set according to the needs, and the mixed excitation experiment is gradually carried out in the simulation operation process. This may include adjusting parameters of the motor's field strength, current, voltage, etc. to change the motor's operating characteristics. The purpose of the hybrid excitation is to adjust the magnetic field during operation of the motor to obtain the desired performance, and to analyze the collected power operation data, including torque and rotational speed data. And using the result of the mixed excitation experiment to know the performance response of the motor, and verifying the consistency of the simulation data and the actual data.

in embodiments of the present invention, the electromagnetic dependent energy efficiency analysis formula is used to evaluate the relationship between the performance of the motor and the magnetic field characteristics. This formula may be research or engineering domain specific and typically includes mathematical relationships between torque, rotational speed and magnetic field parameters, and the magnetic field dependency data for each time point is calculated using an electromagnetic dependency energy efficiency analysis formula. This can be achieved by operating the output torque and rotational speed data with the simulated magnetic field distribution data, and setting a historical magnetic field dependence threshold value in advance, which will be used for subsequent data comparison. The selection of the threshold may depend on the application and performance requirements, comparing the calculated magnetic field dependency data with a historical magnetic field dependency threshold. From the comparison, the marker data points are of one of two types: stabilizing magnetic field data: if the magnetic field dependency data is greater than or equal to the historical magnetic field dependency threshold, the corresponding data point is marked as stable magnetic field data. This means that the operation of the motor under these conditions is stable; unstable magnetic field data: if the magnetic field dependency data is less than the historical magnetic field dependency threshold, the corresponding data point is marked as unstable magnetic field data. This may indicate that the motor is not operating stably under these conditions and may require further adjustment or maintenance.

In the embodiment of the present invention, the unstable magnetic field data obtained in step S25 is used as input. These data contain motor performance information under unstable magnetic field conditions, and appropriate mathematical models of hysteresis effect are selected based on knowledge and experience in the field. This may include some hysteresis loop model (e.g., jiles-Atherton model, etc.) or other mathematical expression relating to hysteresis effects, using the unstable magnetic field data to fit parameters of the hysteresis effect model. This can be achieved by a least squares method or other fitting technique, once the model parameters are fitted, a mathematical model describing the hysteresis effect will be obtained, the obtained steady magnetic field data is used as input, the built mathematical model of hysteresis effect is used to import the steady magnetic field data into the model, the model will return hysteresis bias values, which represent the effect of the hysteresis effect on the motor performance under steady magnetic field conditions, the output torque data, rotational speed data, and hysteresis bias data of the levitation motor obtained in step S25 are used, and an appropriate hybrid excitation mathematical model is selected based on domain knowledge and experience. This model will describe the performance of the motor taking into account hysteresis, using the output torque data, the rotational speed data and the hysteresis deviation data of the levitation motor, fitting the parameters of the hybrid excitation model, once the model parameters are fitted, a mathematical model describing the performance of the motor taking into account hysteresis will be obtained.

Preferably, the electromagnetic dependent energy efficiency analysis formula in step S25 is specifically as follows:

;

wherein eta is expressed as an energy efficiency index of the motor, V is expressed as a three-dimensional space range inside the motor, B is expressed as magnetic induction intensity, mu is expressed as magnetic permeability, H is expressed as magnetic field intensity, J is expressed as distribution density of current over a unit area, E is expressed as electric field intensity,expressed as electromagnetic dependent energy efficiency analysis anomaly adjustment values.

The invention constructs an electromagnetic dependent energy efficiency analysis formula, the main principle of the formula is that in a molecular part, the formula calculates the magnetic field energy in the volume V of the internal space of the motor, and the magnetic field energy is expressed as. Where B is the magnetic induction, H is the magnetic field strength, and μ is the magnetic permeability. This part represents the contribution of the magnetic field to the energy efficiency of the motor, since the magnitude of the magnetic field energy is related to the strength of the magnetic field, and in the denominator part the formula calculates the electric field energy in the volume V of the interior space of the motor, denoted j·e. Where J is the current density and E is the electric field strength. This section shows the contribution of the electric field to the energy efficiency of the motor, since the magnitude of the electric field energy is related to the strength and current density of the electric field, and the energy efficiency index η of the motor can be obtained by calculating the ratio of the magnetic field energy to the electric field energy. If the value of η is high, it means that the magnetic field contributes more to the energy efficiency of the motor and the electric field contributes less to the energy efficiency. Conversely, if the value of η is low, it means that the electric field contributes more to the energy efficiency of the motor, while the magnetic field contributes less to the energy efficiency, according to the three-dimensional empty inside the motor The relationship between the ranges and the parameters constitutes a functional relationship:

;

the magnetic field strength is represented by magnetic induction B and magnetic field strength H, whose changes directly affect the magnitude of the magnetic field energy. By calculating the magnetic field energy, the influence of the magnetic field on the motor energy efficiency can be evaluated, so that the motor design and the magnetic field distribution are optimized, the current density J and the electric field strength E represent the strength of the electric field and the distribution condition of the current in unit area, and the change of the current density J and the electric field strength E directly influence the electric field energy. By calculating the energy of the electric field, the influence of the electric field on the energy efficiency of the motor can be evaluated, so that the motor design and the electric field distribution are optimized, and the space volume V represents the three-dimensional space range inside the motor and determines the energy calculation range. By integrating the whole space volume, the contribution of the magnetic field and the electric field at each position inside the motor is comprehensively considered, the energy efficiency performance of the motor is more comprehensively estimated, and the abnormal adjustment value is analyzed through electromagnetic dependent energy efficiencyFor correcting errors and deviations due to the complexity and non-idealities of the actual system. The method can correct the difference between theoretical assumption in the formula and an actual system, improves the accuracy and reliability of electromagnetic dependent energy efficiency analysis, generates the energy efficiency index eta of the motor more accurately, and simultaneously adjusts the distribution density, magnetic permeability and other parameters of the current in the formula on a unit area according to actual conditions, thereby adapting to different electromagnetic dependent energy efficiency analysis scenes and improving the applicability and flexibility of the algorithm. When the electromagnetic energy efficiency analysis formula conventional in the art is used, the energy efficiency index of the motor can be obtained, and the energy efficiency index of the motor can be calculated more accurately by applying the electromagnetic dependent energy efficiency analysis formula provided by the invention. The formula can improve the energy efficiency performance of the motor, reduce the energy loss and improve the efficiency and the performance of the system by analyzing and optimizing the parameters. In addition, the formula can also detect motor simulation The magnetic field dependence of the magnetic field distribution data helps to find potential magnetic field distribution problems and guides the magnetic field design and optimization process.

Preferably, step S24 comprises the steps of:

step S241: winding analysis is carried out on the magnetic suspension motor according to the motor simulated magnetic field distribution data to obtain stator winding data and exciting winding data; data integration is carried out on stator winding data and exciting winding data, and the stator winding data and the exciting winding data are marked as stationary phase winding data;

step S242: carrying out energization simulation on the magnetic levitation motor, and carrying out magnetic field data acquisition on the magnetic levitation motor according to the winding data in the stationary phase to obtain stator winding magnetic field data and exciting winding magnetic field data in the motor starting phase; performing magnetic field visualization on the stator winding magnetic field data and the exciting winding magnetic field data to generate a stator winding magnetic field influence profile and an exciting winding magnetic field influence profile;

step S243: carrying out distribution path analysis on a motor rotor according to the stator winding magnetic field influence distribution diagram and the exciting winding magnetic field influence distribution diagram to obtain stator winding magnetic field path data and exciting winding magnetic field path data; performing difference analysis through the stator winding magnetic field path data and the exciting winding magnetic field path data to obtain winding magnetic field difference data; performing magnetic field range analysis on the permanent magnet of the levitation motor according to the winding magnetic field difference value data to obtain permanent magnet winding magnetic field data;

Step S244: performing magnetic field intersection detection on the rotor of the levitation motor through the magnetic field data of the permanent magnet winding and the magnetic field data of the exciting winding, so as to obtain mixed exciting magnetic field data; and acquiring current data of the levitation motor by utilizing the electric force sensor based on the mixed excitation magnetic field data to obtain output torque data of the levitation motor and rotating speed data of the levitation motor.

According to the invention, stator winding data and exciting winding data can be obtained by simulating magnetic field distribution data through the motor. This will help to understand the construction and composition of the motor, provide a basis for subsequent analysis by which field data for the stator windings and field windings of the motor during the start-up phase can be obtained through power-on simulation. The data can be used for visualization to generate magnetic field influence patterns of the stator winding and the exciting winding, help to know the magnetic field distribution condition of the motor, and can analyze the distribution path of the motor rotor by utilizing the magnetic field influence patterns of the stator winding and the magnetic field influence patterns of the exciting winding. By means of the difference analysis, winding magnetic field difference data can be obtained, which is helpful for understanding the magnetic field change condition in the motor, and especially in the starting stage, the winding magnetic field difference data can be used for carrying out magnetic field range analysis so as to know the magnetic field data of the permanent magnet winding. This is important for understanding the performance and impact of permanent magnets, and by field intersection detection, hybrid excitation field data can be obtained, which will take into account interactions between the various parts of the motor. The current data of the levitation motor can be acquired by using the electric force sensor, so that output torque data and rotation speed data are acquired. These data will help evaluate the performance of the motor.

As an example of the present invention, referring to fig. 3, the step S24 in this example includes:

in the embodiment of the invention, by preparing motor simulation magnetic field distribution data, which may include geometric parameters of the motor, numerical simulation results of the magnetic field distribution, and the like, specialized electromagnetic simulation software or calculation tools are used, and the tools allow analysis of the magnetic field distribution of the motor and extraction of information about windings, and in the simulation tools, analysis is performed on stator windings. This includes knowledge of the winding pattern of the windings, the geometry of the coils, the material properties of the wires, etc. Data about the stator windings, such as the number of coils, turns, wire cross-sectional area, etc., can be extracted and the excitation windings analyzed in the simulation tool as well. This may include understanding the layout of the exciter coil, material properties, and its interaction with the motor. Data about the field winding, such as the number of coils, turns, wire cross-sectional area, etc., is extracted and the stator winding data and field winding data are integrated together to create a comprehensive data set containing data about the stationary phase winding. This may require unifying the data formats of both for subsequent analysis, and for clarity of distinction, the integrated data is labeled "stationary phase winding data" for explicit use of the data in subsequent steps.

in the embodiment of the invention, the starting stage of the motor is simulated by electrifying through a simulation tool or a motor control system. In this process, current will flow through the stator windings and the field windings, creating a magnetic field, and magnetic field sensors are arranged around the motor to collect magnetic field data. These sensors may be magnetometers, hall effect sensors or magnetic field detection coils, etc., ensuring that the sensor is well connected to the data acquisition device and setting the appropriate sampling rate and data acquisition parameters. The motor is started to simulate its start-up phase operation. In this process, the magnetic field sensor will capture magnetic field data and record the acquired data, including time stamps, location information (if applicable) and magnetic field strength data. And processing and arranging the acquired magnetic field data by using a data analysis tool to obtain magnetic field influence data of the stator winding and the exciting winding, converting the processed data into a graphical magnetic field distribution diagram by using professional magnetic field visualization software such as magnetic field simulation software or data visualization tool, and respectively creating the magnetic field influence distribution diagram for the stator winding and the exciting winding to display the distribution situation of the magnetic field. These figures generally show the strength, direction and spatial distribution of the magnetic field. The acquired and visualized data, including information about the equipment, parameters, and sampling frequency used, are recorded and the generated magnetic field impact profile is saved as a document or report for subsequent analysis and reference.

in the embodiment of the invention, the path of the magnetic field, namely the magnetic field conduction path from one part of the winding to the other part, is identified according to the magnetic field influence distribution diagram of the stator winding. The detailed information of the paths including the shape, length, intensity distribution, etc. of the paths is recorded, and likewise, the conduction path of the magnetic field is analyzed according to the field influence distribution diagram of the exciting winding, and the relevant information of the paths is recorded, and the stator winding field path data is compared with the exciting winding field path data to perform difference analysis. This may include calculating the difference between the two to determine the change in magnetic field around the rotor. This step helps to understand how the field winding affects the field of the stator winding, and uses the difference data to analyze the field range of the permanent magnet. This involves determining the distribution of the magnetic field on the permanent magnet, including intensity, direction and extent. This can be done by means of an analogue or numerical analysis tool, which evaluates the effect of the permanent magnet on the motor performance based on the magnetic field range analysis. This may include evaluating the stability of the permanent magnet, whether the magnetic field distribution meets design requirements, interactions with other components, etc. Recording the results of the distribution path analysis and the difference analysis, including detailed information of the path data, the difference data and the magnetic field range analysis.

In the embodiment of the invention, whether the magnetic fields intersect or not is detected by comparing and analyzing the magnetic field data of the permanent magnet winding and the magnetic field data of the exciting winding. The intersecting area represents the mixed excitation magnetic field data, and the electric force sensors are arranged on the current input line of the motor and are used for monitoring the current of the motor, and the electric force sensors are used for collecting the current data of the levitation motor. Such data may be alternating current or direct current, depending on the design and application of the motor. And correlating the mixed excitation magnetic field data obtained by the magnetic field intersection detection with the current data. This may be achieved by time stamping or other synchronization methods to ensure that the magnetic field data and the current data are time-wise corresponding, and analysis of the collected current data may use signal processing techniques, spectral analysis, etc. to obtain output torque data and rotational speed data of the levitation motor. And recording the mixed excitation magnetic field data, the current data, the output torque data and the rotating speed data. Including time stamp, magnetic field strength, current value, etc.

Preferably, step S3 comprises the steps of:

step S31: noise detection is carried out on the suspension motor through the output torque data of the suspension motor and the rotating speed data of the suspension motor, and motor operation noise data is generated;

step S32: sensor deployment is carried out on the suspension motor according to the motor operation noise data, and environmental data are collected, so that first environmental collection data are obtained;

step S33: performing environmental noise visualization on motor operation noise data based on the first environmental acquisition data to generate an environmental noise change map; the curve curvature screening is carried out on the environmental noise change image, so that an extreme value curvature curve is obtained; confirming coordinate points of the polar bending curve through a homogeneous coordinate system to generate extreme point coordinate data;

step S34: position adjustment is carried out on the sensor according to the extreme point coordinate data, and environmental data are acquired again, so that second environmental acquisition data are obtained; performing interference compensation analysis on the first environment acquisition data and the second environment acquisition data by using an operation vibration interference analysis formula to obtain an operation interference compensation coefficient;

step S35: and carrying out axial phase separation on the suspension motor according to the operation interference compensation coefficient, the suspension motor output torque data and the suspension motor rotating speed data to obtain an axial phase separation mathematical model.

According to the invention, noise detection is carried out through the output torque data and the rotation speed data of the suspension motor, so that the motor operation noise data is generated, the noise level of the motor during operation can be known, and the fault detection and the performance evaluation can be facilitated. Based on motor operation noise data, a sensor is deployed and environmental data is acquired to obtain first environmental acquisition data, influence of environmental factors on motor performance and noise is facilitated to be determined, working environment of the motor can be better understood, the first environmental acquisition data is used for visualizing environmental noise change, extreme point coordinates are identified through analysis of an extreme value curvature curve, distribution and change of environmental noise can be visualized, and meanwhile, further analysis and correction can be facilitated by determining extreme point. Based on the extreme point coordinate data, the position of the sensor is adjusted and environmental data is acquired again to obtain second environmental acquisition data, and the relationship between environmental noise and motor performance can be known more accurately by calibrating the position of the sensor and obtaining second round data. And performing axial phase separation on the motor by using the operation interference compensation coefficient, the output torque data of the suspension motor and the rotation speed data to obtain an axial phase separation mathematical model, better understanding the performance of the motor by considering the interference factor during operation, and establishing the mathematical model so as to improve the control and optimization of the motor behavior.

As an example of the present invention, referring to fig. 4, the step S3 in this example includes:

the invention is arranged on the suspension motor by installing a torque sensor and a rotating speed sensor. The sensors are used for monitoring the torque and the rotating speed of the motor in real time and are connected with the data acquisition system to acquire torque and rotating speed data. These data may be recorded digitally, typically in a number of samples per second, and the acquired data is filtered to remove high or low frequency noise and to ensure data quality. And carrying out noise analysis by using the collected torque and rotating speed data. This may include time and frequency domain analysis to detect the characteristics and spectrum of the noise, and based on the results of the noise analysis, generate noise data as an output of the motor operation. The data may include information such as noise amplitude, spectral content, and time domain waveforms, and the generated noise data is recorded for later analysis and use. This may be in the form of a digital file or a record in a database.

in an embodiment of the invention, the environment parameters related to the motor operation noise are measured by selecting the proper sensor. These sensors may include acoustic sensors, vibration sensors, temperature sensors, humidity sensors, etc., with the specific choice depending on the environmental parameters of interest. For example, acoustic sensors may be used to measure noise levels, vibration sensors may be used to detect vibrations, temperature and humidity sensors may be used to monitor temperature and humidity changes, and based on analysis of motor operating noise data, the optimal position and number of sensors may be determined. The location where the sensor is deployed should take into account factors that may affect motor performance and noise. This may involve mounting the sensor at different locations around the motor, such as the motor housing, bearings, fans, etc. Ensuring that the sensor location captures environmental information related to motor operating noise, installing and connecting the sensor to the data acquisition system. This may require the use of amplifiers, data acquisition cards or sensor modules, with appropriate connections depending on the type and number of sensors. The data acquisition system is configured for continuous data acquisition. It is ensured that the data acquisition system is able to capture environmental data at a sufficient sampling rate for subsequent analysis, to begin recording the environmental data, while operating the levitation motor. Data is ensured to be collected during motor operation to capture environmental changes associated with motor noise. The data is continuously recorded to ensure that enough samples are obtained for analysis, resulting in first environmental acquisition data.

in the embodiment of the invention, the data acquired by the first environment is correlated with the motor operation noise data by using a proper data analysis tool. This may involve time synchronizing the environmental data with the motor noise data for comparative analysis, and plotting the environmental noise change map using a visualization tool, such as a chart or plotting software. In general, the horizontal axis represents time, the vertical axis represents the intensity or amplitude of the environmental noise, and the curvature of the noise curve is observed and analyzed on the environmental noise change map. The curvature represents the rate of change of the curve and can be calculated by derivative or difference. A high degree of curvature may represent a significant change in ambient noise and suitable algorithms or methods are used to identify curved portions on the curve, which may contain extreme points, from which the extreme curvature curve is generated as a result of curve curvature screening. This is a curve showing significant extremum in the ambient noise variation, converting the extremum camber curve into a homogeneous coordinate system for easier identification of the extremum points. The homogeneous coordinate system is generally used for analyzing the bending curve, and can help identify special points of the curve. These points generally correspond to significant events or changes in the ambient noise data. And recording coordinate data of the extreme points, including information such as time, noise intensity and the like. These data will be the basis for subsequent analysis to understand the correlation between ambient noise and motor operating noise.

in the embodiment of the present invention, the relationship between the current sensor position and the environmental noise generating source is evaluated according to the extreme point coordinate data obtained in step S33, and the sensor position is adjusted to minimize the interference of the environmental noise. This may require repositioning or reinstalling the sensor closer to the operating device of interest, such as a motor, ensuring that the sensor position is adjusted to accurately collect environmental data and correlate with the motor operating data, and re-collecting the environmental data after the sensor position adjustment is completed. And ensuring that the acquired data comprise environmental noise information during the operation of the motor, and comparing and analyzing the first environmental acquired data and the second environmental acquired data by utilizing an operation vibration interference analysis formula. This may involve a comparison of vibration frequency, amplitude and vibration pattern, and from the analysis, the correlation and degree of interference between ambient noise and motor operating vibrations is identified. And calculating the operation interference compensation coefficient based on the result of the comparison analysis. This coefficient can be used to correct for interference of ambient noise with motor operating data, thereby improving accuracy and reliability of the data.

Preferably, the operation vibration disturbance analysis formula in step S34 is specifically as follows:

；

where G (ω, T) is expressed as a vibration disturbance response coefficient varying with time T and frequency ω, a is expressed as a maximum amplitude of vibration disturbance, α is expressed as a time attenuation coefficient, β is expressed as a vibration frequency, ω is expressed as an initial phase shift, V is expressed as a weight affecting an integrated response amplitude, C is expressed as a weight in a nonlinear relation with frequency and time, δ is expressed as a weight controlling a nonlinear attenuation speed, T is expressed as an end time of integration, and σ is expressed as an operational vibration disturbance analysis abnormal correction amount.

The invention constructs an operation vibration interference analysis formula, wherein the maximum amplitude A of vibration interference represents the amplitude of vibration interference, the larger A value represents stronger vibration interference, the smaller A value represents weaker vibration interference, and the time attenuation coefficient alpha controls the speed of vibration interference attenuation along with time. A larger value of a will result in a faster damping of the vibration disturbance, while a smaller value of a will result in a slower damping of the vibration disturbance, the vibration frequency β representing the frequency of the vibration. When the beta value is larger, the frequency of vibration is higher; when the value of β is small, the frequency of the vibration is low, and the initial phase shift ω represents the initial phase shift of the vibration. Different ω values will cause a change in the phase of the vibration, which constitutes a functional relationship based on the maximum amplitude of the vibration disturbance and the correlation between the above parameters:

;

By adjusting parameters A, alpha, beta, omega and the like, vibration interference of different types and intensities can be simulated so as to better understand and analyze vibration interference phenomena in an actual system, by adjusting parameters alpha and T, attenuation trend of the vibration interference along with time can be observed, the stability of the system and influence of the vibration interference on the system performance can be evaluated, by adjusting parameters C and delta, nonlinear relation between frequency and time can be simulated and analyzed, nonlinear characteristics possibly existing in the system are revealed, amplitude of integral response can be controlled by adjusting parameters V, so that influence of the vibration interference can be better analyzed and understood, and by adjusting operation vibration interference analysis abnormal correction quantity sigma, errors and deviations caused by complexity and nonideal of the actual system are corrected. The method can correct the difference between theoretical assumption in the formula and an actual system, improves the accuracy and reliability of operation vibration interference analysis, generates vibration interference response coefficient G (omega, t) changing along with time t and frequency omega more accurately, and meanwhile, parameters such as vibration frequency, weight for controlling nonlinear attenuation speed and the like in the formula can be adjusted according to actual conditions, so that the method is suitable for different operation vibration interference analysis scenes, and improves the applicability and flexibility of the algorithm. When the operation vibration analysis formula conventional in the art is used, the vibration interference response coefficient changing along with time t and frequency omega can be obtained, and the vibration interference response coefficient changing along with time t and frequency omega can be calculated more accurately by applying the operation vibration analysis formula provided by the invention. The formula can simulate, analyze and correct the influence of vibration interference by adjusting the functions among different parameters, help understand and optimize the vibration performance of the system, and accurately evaluate the influence of the vibration interference on the operation of the system.

Preferably, step S35 includes the steps of:

step S351: performing operation time sequence analysis on the output torque data of the suspension motor and the rotating speed data of the suspension motor to obtain operation time sequence analysis data;

step S352: carrying out peak value data extraction on output torque data of the suspension motor and rotation speed data of the suspension motor by utilizing the operation time sequence analysis data to generate output peak value torque data and motor peak value rotation speed data;

step S353: the method comprises the steps of performing displacement monitoring on a suspension motor rotor through output peak torque data and motor peak rotating speed data to generate suspension motor rotor displacement data; mirror image data division is carried out on the suspension motor rotor displacement data to obtain upper tooth pitch displacement data and lower tooth pitch displacement data;

step S354: performing axial displacement analysis on the upper tooth pitch displacement data and the lower tooth pitch displacement data to obtain rotor axial displacement data; position tracking is carried out on the rotor of the suspension motor through the rotor axial displacement data, so that real-time rotor displacement feedback data are obtained;

step S355: comparing the tooth pitch positions according to the phase-splitting algorithm to the real-time rotor displacement feedback data to obtain tooth pitch position deviation data; comparing the tooth pitch position deviation data with a preset tooth pitch position deviation threshold value, and when the tooth pitch position deviation data is larger than the tooth pitch position deviation threshold value, performing clockwise moment adjustment on the suspension motor according to the tooth pitch position deviation data to generate clockwise adjustment data; when the tooth pitch position deviation data is smaller than the tooth pitch position deviation threshold value, carrying out counterclockwise moment adjustment on the suspension motor according to the tooth pitch position deviation data to generate counterclockwise adjustment data;

Step S356: performing axial split-phase control based on the clockwise adjustment data or the anticlockwise adjustment data to obtain axial split-phase adjustment data; and performing data fitting on the axial split-phase adjustment data by using the operation interference compensation coefficient so as to generate an axial split-phase mathematical model.

The invention obtains detailed information about the operation of the motor by carrying out time sequence analysis on the output torque and the rotating speed data of the suspension motor. Such data may include the operating state, run time, and other dynamic characteristics of the motor. By analyzing the data, the performance of the motor can be better known, and the output torque and rotation speed data of the levitation motor can be analyzed in a time sequence to obtain detailed information about the operation of the motor. Such data may include the operating state, run time, and other dynamic characteristics of the motor. By analyzing these data, the performance of the motor can be better understood. By using the peak torque and rotational speed data, the displacement of the suspended motor rotor can be monitored. These displacement data can help to understand the operation of the motor inside. The displacement data is further divided into upper and lower pitch displacement data to provide more detailed information. Axial displacement data of the rotor can be obtained by axially analyzing the displacement data. This data can be used to track the real-time position of the motor rotor, which is important for control and monitoring. And comparing the real-time rotor displacement data with a preset tooth pitch position deviation. If the deviation of the tooth pitch position is larger than a preset threshold value, corresponding moment adjustment data are generated to adjust the operation of the suspension motor. This helps to maintain the stability and performance of the motor. And comparing the real-time rotor displacement data with a preset tooth pitch position deviation. If the deviation of the tooth pitch position is larger than a preset threshold value, corresponding moment adjustment data are generated to adjust the operation of the suspension motor. This helps to maintain the stability and performance of the motor, to improve the efficiency of the motor, to reduce vibration and to extend the life of the motor.

In the embodiment of the invention, the output torque data and the rotation speed data of the suspension motor are collected by using a proper sensor, the data are collected within a certain time period, the continuity and the integrity of the data are ensured, a time sequence analysis tool (such as an FFT or sliding window method) is utilized to carry out time sequence analysis on the collected data, relevant characteristics or modes are extracted from the time sequence analysis tool, and operation time sequence analysis data are generated. And (3) finding out peak values of output torque and rotating speed from time sequence analysis data, recording the peak value data for later steps, monitoring the displacement of the rotor of the suspension motor by using sensor data, and dividing the displacement into upper tooth pitch displacement data and lower tooth pitch displacement data according to the monitoring data. And analyzing the displacement data, particularly the displacement in the axial direction, tracking the position of the rotor in real time by utilizing the data, generating real-time rotor displacement feedback data, comparing the real-time rotor displacement data with the expected tooth pitch position by using a split-phase algorithm, and adjusting the torque of the suspension motor to correct the deviation if the deviation is detected to exceed a preset threshold value. Depending on the direction of the deviation, it is decided whether to adjust clockwise or counterclockwise. And (3) according to the adjustment data of the previous step, performing axial split-phase control, performing data fitting on the adjustment data by utilizing the operation interference compensation coefficient, and finally generating an axial split-phase mathematical model.

Preferably, step S4 comprises the steps of:

step S41: leading the mixed excitation mathematical model and the axial split-phase mathematical model into a motor shell three-dimensional model for internal space modeling to generate a suspended motor internal model;

step S42: model coupling is carried out on the suspended motor internal model and the motor shell three-dimensional model, and a three-dimensional coupling model is generated;

step S43: performing simulated operation motor data acquisition on the three-dimensional coupling model to obtain motor performance mapping data;

step S44: performing simulated operation environment data acquisition on the three-dimensional coupling model to obtain multi-physical-field interaction data;

step S45: and carrying out motor component interconnection analysis on the three-dimensional coupling model based on the motor performance mapping data and the multi-physical-field interaction data to generate motor component interconnection data.

According to the invention, the mixed excitation mathematical model and the axial split-phase mathematical model are led into the three-dimensional model of the motor shell to perform internal space modeling, so that the internal model of the suspension motor is generated, and the mixed excitation mathematical model and the axial split-phase mathematical model are used for establishing the mathematical model of the internal structure of the motor. These models may include electromagnetic properties, mechanical properties, control systems, etc. of the motor. These models are imported into a three-dimensional housing model of the motor to create an interior space model of the motor. The internal model is coupled to the housing model to create a complete three-dimensional model that contains the external and internal structures of the motor. This integrated model facilitates more comprehensive motor analysis. The operation of the motor is simulated using a three-dimensional coupling model to obtain data on the performance of the motor. This may include the output torque, efficiency, power characteristics, etc. of the motor. These data help to understand the performance of the motor. And simulating the operation of the motor under different environmental conditions so as to acquire multi-physical-field interaction data. Such data may include information on temperature, vibration, electromagnetic fields, etc., which may be helpful in assessing the performance of the motor in different environments. And analyzing the performance data and the multi-physical field data of the motor to know the mutual influence and interconnection relation among all the components of the motor. This helps to improve the design and performance of the motor and provides useful information about the interactions between the various components of the motor.

In the embodiment of the invention, the mixed excitation mathematical model and the axial split-phase mathematical model are led into a motor shell three-dimensional model of the suspension motor to model the internal space. Particular embodiments may include: the three-dimensional geometric model of the motor housing is built using specialized modeling software, such as Computer Aided Design (CAD) tools, integrating the hybrid excitation mathematical model, which may include mathematical models that take into account electromagnetic fields, current distribution, flux distribution, etc., integrating the axial split-phase mathematical model, which may involve mechanical modeling, such as taking into account mathematical descriptions of the rotor, stator, bearings, etc. Once the interior and exterior shell models of the levitation motor are generated, the two are then model coupled to generate an integrated three-dimensional coupling model. This process may include: the internal model is combined with the shell model in a CAD or other simulation tool to ensure that they are fully coupled, and the interfaces and connections of the shell and the interior are designed in detail to ensure that they are seamlessly connected. And performing simulation operation on the motor by using the three-dimensional coupling model to acquire motor performance mapping data. Particular embodiments may include: setting simulation running conditions such as current input, load condition and the like, running simulation and recording motor performance data such as output torque, speed, efficiency and the like, and running for a plurality of times under different running conditions so as to obtain the diversity of the performance mapping data. The three-dimensional coupling model is simulated to acquire multi-physical-field interaction data so as to know the behaviors of the motor under different environmental conditions. Particular embodiments may include: various environmental conditions, such as different temperatures, humidity, vibrations, etc., are simulated, and data related to these environmental conditions, such as temperature distribution, vibration response, electromagnetic field distribution, etc., are collected. And carrying out interconnection analysis among the motor parts according to the motor performance mapping data and the multi-physical-field interaction data so as to generate motor part interconnection data. Particular embodiments may include: interactions between individual components of the motor are studied using data analysis tools and simulation results to generate motor component interconnection data to clearly express these interactions and interconnection relationships.

Preferably, step S5 comprises the steps of:

step S51: carrying out distributed processing on motor component interconnection data by utilizing a cloud computing platform so as to generate distributed operation processing data;

step S52: historical data collection is carried out on the distributed operation processing data to obtain historical distributed operation processing data; carrying out data set division on the historical distributed operation processing data to generate a model training set and a model testing set; model training is carried out on the model training set based on a support vector machine algorithm, and a motor performance prediction model is generated; performing model test on the motor performance prediction model by using a model test set to obtain performance prediction result data;

step S53: and performing control optimization and visualization on the three-dimensional coupling model according to the performance prediction result data, so as to generate an axial split-phase hybrid excitation type magnetic suspension motor modeling scheme.

According to the invention, the motor component interconnection data is processed in a distributed manner by utilizing the cloud computing platform. This means that the data processing task is broken down into multiple parallel tasks that can be executed simultaneously on the cloud server to speed up processing and improve efficiency, and the historical distributed running process data is collected and separated into a model training set and a model test set. The model training set is then trained using a Support Vector Machine (SVM) or other machine learning algorithm to generate a motor performance prediction model, the data is divided into a training set and a test set to verify the performance of the model, the model training is performed using a support vector machine or other suitable machine learning algorithm to predict motor performance, and the performance of the model is evaluated using the model test set to ensure its accuracy and reliability. Based on the performance prediction result data, optimizing operation and control strategies of the motor to improve performance and efficiency, and utilizing a visualization tool to present the optimized model in an intuitive manner so that engineers and decision makers can better understand and analyze characteristics of the model, and generating a modeling scheme of the axial split-phase hybrid excitation type magnetic suspension motor according to the optimized model, wherein the modeling scheme comprises design parameters, control strategies and the like.

In the embodiment of the invention, through acquiring interconnection data related to motor components, which may include various parameters such as sensor data, operation records, temperature, humidity and the like, cleaning and preprocessing the data to remove error data, process missing values and adapt the data to carry out subsequent analysis, and applying statistical and machine learning techniques to analyze the data to find potential modes, correlations and trends, and generating distributed operation processing data according to the result of data analysis, which is used for subsequent modeling and optimization steps. The historical distributed operating process data is used to build a motor performance prediction model, comprising the following processes: historical distributed operation processing data is obtained from step S51, and the historical data is divided into two parts, one for model training and the other for model testing. Typically, dataset partitioning is to evaluate the performance and generalization ability of a model, employing a Support Vector Machine (SVM) algorithm or other machine learning algorithm, using a model training set to construct a motor performance prediction model. During training, the model learns how to predict motor performance from the input data, and the performance of the model is evaluated using the model test set, which helps determine whether the model can accurately predict motor performance. And optimizing a control system of the motor according to the performance prediction result data to improve performance, efficiency and reliability. This may involve adjusting control parameters, feedback control strategies, etc., using a visualization tool to visually present the motor model and the optimized control strategy. This helps engineers and decision makers to better understand the characteristics of the model and optimize the impact on motor performance.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The modeling method of the axial split-phase mixed excitation type magnetic suspension motor is characterized by comprising the following steps of:

2. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 1, wherein the step S1 comprises the steps of:

step S11: acquiring a multi-view magnetic levitation motor image set;

3. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 2, wherein the step S2 comprises the steps of:

step S21: acquiring data of a magnetic suspension motor;

4. A modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 3, wherein the electromagnetic dependent energy efficiency analysis formula in step S25 is as follows:

;

5. A method of modeling an axial split-phase hybrid excitation type magnetic levitation motor according to claim 3, wherein step S24 comprises the steps of:

6. A method of modeling an axial split-phase hybrid excitation type magnetic levitation motor according to claim 3, wherein step S3 comprises the steps of:

7. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 6, wherein the operation vibration disturbance analysis formula in step S34 is as follows:

;

8. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 6, wherein the step S35 comprises the steps of:

9. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 7, wherein the step S4 comprises the steps of:

10. The modeling method of an axial split-phase hybrid excitation type magnetic levitation motor according to claim 7, wherein the step S5 comprises the steps of: