Permanent magnet synchronous motor health assessment and early warning method and system based on digital twinning

Abstract

The invention relates to a permanent magnet synchronous motor health assessment and grading health early warning method and system based on digital twinning, and the method comprises the steps: constructing a digital twinning model consistent with the actual operation mechanism of a motor, carrying out the fusion analysis of entity operation data and virtual prediction data, and achieving the quantitative assessment of the operation health state of the motor. According to the method, multi-source operation parameters of a motor are collected in real time, a dynamically updated entity operation database is established, a multi-physics field coupled simulation model is constructed, and quick real-time response of a virtual model is achieved through partitioned model order reduction processing. And on the basis, a comprehensive health index is constructed based on virtual-real consistency evaluation, an entity operation state evaluation result and a virtual prediction deviation evaluation result are fused, and grading health judgment and early warning response of the motor operation state are realized according to the comprehensive health index. According to the invention, the accuracy and stability of motor health assessment can be improved, the risk of false alarm and missing alarm is reduced, and the method is suitable for on-line health monitoring and predictive maintenance of the permanent magnet synchronous motor.

Description

Technical Field

[0001] This invention relates to the field of motor condition monitoring and health assessment technology, specifically to a method and system for health assessment and early warning of permanent magnet synchronous motors based on digital twins. Background Technology

[0002] As a key actuator in modern industrial equipment, new energy systems, and various power devices, the operating status of electric motors directly affects the safety, reliability, and efficiency of the system. Therefore, continuous monitoring of motor operating status and early identification of potential anomalies or failure risks are essential for achieving safe equipment operation and predictive maintenance.

[0003] Existing motor condition monitoring technologies typically rely on placing sensors on the motor body or external structure to collect operating parameters such as current, voltage, temperature, and vibration to analyze and judge the motor's condition. These methods are primarily based on externally measurable signals for evaluation. However, limitations imposed by sensor placement, motor dimensions, and operating environment make it difficult to comprehensively reflect changes in the multi-physical field coupling mechanisms within the motor, including electromagnetic, thermal, and structural aspects. Furthermore, the stability and accuracy of monitoring results are somewhat limited under complex operating conditions.

[0004] With the development of digital twin technology, some technical solutions attempt to combine real-time operating data with simulation models by constructing virtual models corresponding to physical motors, thereby achieving dynamic mapping and prediction of motor operating states. This type of method, to some extent, expands the ability of traditional monitoring methods to understand the internal state of motors. However, existing solutions generally suffer from complex model construction, high computational load, and difficulty in meeting the real-time requirements of online operation, thus limiting their application in practical engineering scenarios.

[0005] Furthermore, existing digital twin motor condition assessment schemes mostly focus on the identification or qualitative analysis of specific failure modes, and virtual models are often used as auxiliary analysis tools. They lack a systematic constraint and evaluation mechanism on the consistency between physical operating data and virtual model prediction results, making it difficult to form stable and interpretable operating health assessment results under multiple operating conditions.

[0006] In terms of health status assessment and early warning, existing technologies are usually based on single monitoring thresholds or empirical rules for judgment. They lack a quantitative assessment framework that integrates multi-source entity monitoring information with virtual model prediction information, making it difficult to achieve continuous characterization of the overall operating health status of the motor and to support the hierarchical judgment requirements from early anomaly identification to severe fault early warning.

[0007] Therefore, there is an urgent need for a permanent magnet synchronous motor health assessment and early warning technology solution that can effectively integrate the motor operation mechanism model with real-time monitoring data while ensuring online operation capability, evaluate the consistency between the physical operation state and the virtual predicted state, and transform the assessment results into quantifiable health indicators and hierarchical early warning mechanisms, so as to improve the reliability of motor operation status assessment and fault early warning capability. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for health assessment and early warning of permanent magnet synchronous motors based on digital twins. This invention constructs a digital twin model that is consistent with the actual operating mechanism of permanent magnet synchronous motors and integrates and analyzes physical monitoring data with virtual prediction results. This solves the problems of lagging identification of potential failure risks and high false alarm rate caused by relying solely on physical monitoring data and lacking mechanism consistency constraints in the prior art. Thus, it realizes quantitative assessment and graded early warning of the operating health status of permanent magnet synchronous motors.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: a method for health assessment and early warning of permanent magnet synchronous motors based on digital twins, comprising the following steps:

[0010] Step S1: Collect the operating parameters of the permanent magnet synchronous motor in real time, preprocess the operating parameters, and construct a multi-source entity operation database that is dynamically updated over time.

[0011] Step S2: Construct a digital twin model that is consistent with the actual operating mechanism of the permanent magnet synchronous motor, and obtain virtual predicted operating data corresponding to the operating parameters;

[0012] Step S3: Perform model reduction processing on the digital twin model to transform the high-dimensional multiphysics coupling model into a reduced-order virtual model that can run online, so as to meet the real-time computational requirements of real-time health assessment and early warning analysis.

[0013] Step S4: Based on the virtual prediction operation data of the multi-source entity operation database and the reduced-order virtual model, perform virtual-real consistency assessment and calculate a comprehensive health index, wherein the comprehensive health index is obtained by fusing the baseline health assessment result based on the entity operation status and the abnormal deviation assessment result based on the degree of virtual-real deviation.

[0014] Step S5: Compare the comprehensive health index with the preset threshold range to determine the motor health level, and trigger the corresponding graded health warning response according to the motor health level.

[0015] Furthermore, in step S1, the operating parameters include at least two or more of the following: current, voltage, rotational speed, temperature, and vibration parameters, and further include one or more of the following cooling-related operating parameters: inlet air temperature, air velocity, air volume, and heat exchange boundary characterization parameters.

[0016] Furthermore, in step S1, the preprocessing includes: data cleaning, missing value handling, outlier removal, time alignment, format unification, standardization processing, and storage index management;

[0017] The preprocessing process includes:

[0018] Step S101: Perform preliminary screening on the raw multi-parameter datasets from different data sources, removing erroneous values, missing values, and outlier values;

[0019] Step S102: Convert the cleaned multi-parameter data into a unified data structure and format and complete time alignment;

[0020] Step S103: Perform encoding and compression processing on the standardized data;

[0021] Step S104: Write the encoded and compressed multi-parameter data into the multi-source entity runtime database;

[0022] Step S105: Classify and organize the data and create an index in the multi-source entity running database.

[0023] Further, step S2 specifically includes:

[0024] An electromagnetic solution model is established based on the motor structure and material parameters to obtain electromagnetic force and loss.

[0025] The temperature distribution is obtained by inputting the loss as a heat source into the thermal field model and combining it with heat transfer boundary conditions.

[0026] Electromagnetic force is used as the excitation input to obtain the vibration response of the structural dynamics model, and the equivalent correlation between the vibration response and the radiated noise characteristics is further established to form a digital twin model with multi-physics coupling.

[0027] The digital twin model includes at least two or more of the following: electromagnetic field model, thermal field model, and structural dynamics field model, and is used to reflect the multi-physics field coupling operation mechanism inside the motor.

[0028] Furthermore, step S3 employs a partitioned model order reduction strategy, reducing the order of different physical field sub-models separately and integrating them at the system level, including:

[0029] Acquire multi-condition simulation snapshot data;

[0030] Feature extraction and dimensionality reduction are performed on temperature field snapshot data. A fast prediction model for the temperature field is constructed with operating parameters as input and low-dimensional features as output.

[0031] A low-order model of the structural dynamic response is established, and an equivalent transfer relationship between vibration response and noise characteristics is constructed.

[0032] Configure data interfaces for each reduced-order sub-model and integrate them into a system-level real-time reduced-order virtual model.

[0033] Further, in step S4, the virtual-real consistency assessment includes: calculating the deviation or residual between the measured parameters of at least one type of entity and the virtual predicted parameters, and using the deviation or residual as input to the abnormal deviation assessment result to reflect the degree of deviation of the motor operating state from the mechanism consistency prediction result.

[0034] The fusion weight of the baseline health assessment result and the abnormal deviation assessment result in the comprehensive health index is adaptively adjusted according to the operating conditions. The operating conditions include at least one or more of speed, load rate and cooling conditions, so as to improve the sensitivity to the abnormal deviation assessment results when under high load, high speed or deteriorating cooling conditions.

[0035] Furthermore, in step S5, the motor health level is classified into at least three levels. The corresponding motor health level is output based on the comparison results of the comprehensive health index and at least two preset threshold intervals, and corresponding early warning prompts and handling guidelines are configured for different motor health levels.

[0036] This invention also provides a health assessment and early warning system for permanent magnet synchronous motors based on digital twins, used to implement the above-mentioned method, including:

[0037] The data acquisition module is used to collect motor operating parameters and build a multi-source entity operating database;

[0038] The virtual model module is used to construct a digital twin model that is consistent with the actual operating mechanism of the motor and output virtual predictive operating data. It also includes a reduction unit for model reduction processing to form a reduced virtual model that can run online.

[0039] The data analysis and processing module is used to integrate the multi-source entity operation database and the virtual prediction operation data, perform virtual-real consistency assessment and calculate comprehensive health indicators, and determine the motor health level based on the comprehensive health indicators;

[0040] The visualization health assessment and early warning module is used to display the physical operating status, virtual prediction status and health assessment results, and trigger graded health early warning responses based on the motor health level.

[0041] Furthermore, the data analysis and processing module is configured to perform a virtual-real consistency assessment on the motor entity's operating data and the digital twin model's predicted data, and generate an anomaly deviation assessment result based on the degree of deviation between the entity's operating state and the virtual predicted state.

[0042] The data analysis and processing module is further configured to integrate the baseline health assessment results based on the entity's operating status with the abnormal deviation assessment results to generate a comprehensive health index that characterizes the overall operating health level of the motor.

[0043] Furthermore, the visual health assessment and early warning module is configured to classify and determine the motor operating status based on the comparison results of the comprehensive health indicators and the preset threshold range, and output corresponding level health early warning information.

[0044] The visualized health assessment and early warning module is also configured to collect human-computer interaction feedback information and send the feedback information back to the data analysis and processing module for adaptive optimization of model parameters or health assessment strategies.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] (1) This invention constructs a digital twin model that is consistent with the actual operating mechanism of a permanent magnet synchronous motor, and transforms the multi-physics field coupling simulation model into a lightweight virtual model that can be run online. This enables the virtual model to run in real time while ensuring physical consistency, thus overcoming the problems of large computational load and difficulty in online monitoring of the simulation model in the prior art.

[0047] (2) By introducing a virtual-real consistency health assessment mechanism, this invention integrates the safety assessment based on the physical operating status with the deviation assessment based on the prediction results of the digital twin model, constructs a comprehensive safety coefficient, realizes a unified quantitative representation of the operating status of the permanent magnet synchronous motor, and improves the reliability and stability of anomaly identification under different operating conditions through a weight adaptive adjustment mechanism, thereby reducing the risk of false alarms and missed alarms.

[0048] (3) Based on the comprehensive safety factor, the present invention establishes a graded safety early warning mechanism, realizes full coverage of permanent magnet synchronous motor from early abnormal identification to serious fault early warning, and provides clear early warning response basis for operation and maintenance, thereby improving the safety and maintainability of motor operation. Attached Figure Description

[0049] Figure 1 This is a flowchart of a health assessment and early warning method for permanent magnet synchronous motors based on digital twins provided in an embodiment of the present invention;

[0050] Figure 2This is a schematic diagram of the multi-parameter preprocessing process in an embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of the process of constructing a multiphysics coupling simulation model in an embodiment of the present invention;

[0052] Figure 4 This is a flowchart illustrating the process of reducing the order of a multiphysics coupled simulation model based on a partitioned model reduction method in an embodiment of the present invention.

[0053] Figure 5 This is a flowchart illustrating the comprehensive assessment and graded early warning determination of motor health status in an embodiment of the present invention;

[0054] Figure 6 This is a structural diagram of the health assessment and early warning system for permanent magnet synchronous motors based on digital twins provided in an embodiment of the present invention. Detailed Implementation

[0055] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0056] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0057] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0058] The following detailed description, in conjunction with the accompanying drawings, uses an air-cooled permanent magnet synchronous motor with a rated power of 1.5kW and a rated speed of 1500rpm as an example to illustrate embodiments of the present invention. These embodiments are merely illustrative of the technical solutions of the present invention and do not constitute a limitation on the motor's power rating, speed, or cooling method. In the following detailed description, numerous specific details are set forth to facilitate explanation and provide a comprehensive understanding of the embodiments of the present invention. However, one or more embodiments may be practiced by those skilled in the art without these specific details.

[0059] This embodiment provides a method for health assessment and early warning of permanent magnet synchronous motors based on digital twins, such as... Figure 1As shown, this method achieves a quantitative assessment of the motor's operational health status by fusing and analyzing the actual operating data of the motor with the prediction results of a digital twin model, and triggers a tiered early warning response when health indicators are abnormal. The implementation steps of this method are as follows.

[0060] S1. Real-time acquisition of motor operating parameters: current, voltage, speed, temperature, etc., and preprocessing of the parameter data to construct a multi-source entity operation database that is dynamically updated over time, used to characterize the actual operating status of the motor under different working conditions.

[0061] S2. A detailed 3D model of the motor's internal and external structures is created to provide a unified basic model for simulations of various physical fields. Based on this, a multi-physics coupled simulation model of the permanent magnet synchronous motor is constructed using the finite element analysis method, including electromagnetic field simulation analysis, magnetic-thermal bidirectional coupling simulation analysis, and vibration and noise field simulation analysis. This model is used to obtain a virtual predicted operating state consistent with the actual operating mechanism of the motor and to construct a complete simulation dataset.

[0062] S3. Based on the partitioned model reduction method, the multiphysics coupling simulation model is reduced in order, and the high-dimensional multiphysics simulation model is transformed into a reduced-order virtual model that can run in real time, so as to meet the real-time requirements of online operation and health assessment analysis of digital twin models.

[0063] S4. Based on the simulation analysis results of real-time monitoring data and reduced-order virtual model, the overall operating health status of the motor is evaluated. Through the preset safety performance threshold and the simulation analysis results, the comprehensive safety factor of the permanent magnet synchronous motor is calculated based on the normalized safety factor model. The comprehensive safety factor is used as a health indicator, and a corresponding level of early warning response is triggered when it is lower than the corresponding preset threshold.

[0064] Further, the multi-parameter preprocessing steps are as follows: Figure 2 As shown, the details are as follows:

[0065] S101. Perform preliminary screening on the raw multi-parameter datasets collected from different data sources, and deeply remove erroneous values, missing values ​​and extreme outliers from the data.

[0066] S102. Convert the cleaned multi-parameter data into a unified data structure and format;

[0067] S103. Perform encoding and compression processing on standardized data to reduce storage usage while ensuring data integrity and retrieval.

[0068] S104. Write the encoded and compressed multi-parameter data into a multi-source spatial database to achieve unified storage management;

[0069] S105. Classify and organize the data in the database and create an index to achieve efficient retrieval, so as to support the online operation and health assessment process of the digital twin model.

[0070] Preferably, the arrangement of data acquisition sensors in S101 must not only meet monitoring requirements but also consider economy and miniaturization: Temperature information of the target motor is acquired using different types of temperature sensors and multi-channel temperature acquisition units with communication capabilities, and these sensors are respectively placed at the stator winding ends, the housing surface, and the cooling air duct locations. Key vibration data of the motor is acquired using triaxial accelerometers, and these sensors are placed at key structural locations such as the motor stator housing and drive end cover. Motor speed, current, torque, and other signals are acquired using dynamic torque sensors and matching instruments; the cooling condition-related parameters are used as one of the operating condition inputs for subsequent health assessment and virtual-real consistency analysis; the sensor data is transmitted to the host computer via a communication bus for data preprocessing, and the preprocessed data is sent to the data processing module via network communication.

[0071] Furthermore, the steps for constructing a multiphysics coupled simulation model based on a numerical simulation platform are as follows: Figure 3 As shown, the details are as follows:

[0072] S201. Perform detailed 3D modeling of the internal and external structure of the motor to provide a unified basic model for simulation of various physical fields;

[0073] S202. Set the electromagnetic parameters, material properties and excitation conditions of the motor to form an electromagnetic field solution model;

[0074] S203. Calculate electromagnetic force, loss and electromagnetic heat source to provide input data for subsequent coupling;

[0075] S204. Electromagnetic loss is mapped as a heat source to the temperature field model, and corresponding heat transfer boundary conditions are set according to the motor cooling method. The temperature distribution of the winding, permanent magnet and iron core is simulated. The temperature results are fed back to the electromagnetic model to correct the magnetic properties of the material and realize the magnetic-thermal bidirectional coupling.

[0076] S205. Extract the spatiotemporal distribution of electromagnetic force as an excitation, analyze the vibration response of the computer shell through structural harmonic response, and further obtain the radiated noise characteristics.

[0077] S206. Integrate the various physical field models to establish a multi-physics coupling model with multiple operating condition parameters as inputs and multiple physical field parameters as outputs, and establish a complete simulation dataset.

[0078] Furthermore, methods for reducing the order of partitioned models, such as... Figure 4 As shown, the details are as follows:

[0079] S301. Perform multi-physics bidirectional coupling simulation of the target permanent magnet synchronous motor under different working conditions, obtain complete physical field snapshot data under each working condition, including: electromagnetic field distribution data, temperature field distribution data, structural response data and sound field radiation data, and establish a training sample dataset;

[0080] S302. Extract temperature field snapshot data, and obtain low-dimensional feature quantities that reflect the main changes in the temperature field through feature extraction and dimensionality reduction methods; construct a surrogate model with operating parameters as input and dimensionality-reduced feature coefficients as output to form a lightweight magnetic-thermal bidirectional coupling model that can be solved quickly.

[0081] S303. Perform dynamic modeling of the motor structure, construct a low-order dynamic model that reflects the vibration characteristics of the structure, extract the main harmonic components of the electromagnetic force as excitation, establish the linear transmission relationship between structural vibration and radiated noise, and form a vibration-noise reduced-order model.

[0082] S304. The magnetic-thermal coupling proxy model and the vibration-noise reduction model are encapsulated as standard functional model units. The functional model units are integrated and the data interface is configured in the system-level integration environment to build a complete system-level real-time reduction model.

[0083] Furthermore, in this embodiment, the lightweight magneto-thermal surrogate model and the lightweight vibration and noise model are integrated into the same system through a system-level simulation and integration platform, and hardware-in-the-loop testing is performed. Key time-varying parameters in the reduced-order model are calibrated online through a state estimation and parameter update mechanism to maintain virtual-real consistency and improve the accuracy of health assessment. The aforementioned integration platform, interface standards, and state estimation algorithms can all be selected according to engineering conditions and do not constitute a limitation of this invention.

[0084] Furthermore, the steps for assessing the health status of the target motor are as follows: Figure 5 As shown, its core lies in constructing a comprehensive safety coefficient based on virtual-real consistency, which unifies and quantitatively represents the actual operating state of the motor with the predicted state of the digital twin model, thereby achieving the assessment and early warning judgment of the motor's operating health level. In this embodiment, the comprehensive safety coefficient is obtained by fusing calculations of the baseline safety state and the degree of abnormal deviation.

[0085] In this embodiment, the comprehensive safety factor can be further normalized, and its normalized mathematical expression is:

[0086]

[0087] Here, k is a scaling factor used to adjust the sensitivity of the comprehensive safety factor in the normalization mapping process, and its value can be set according to actual engineering needs. In this implementation, k=6 is taken as an example value.

[0088] The comprehensive safety factor The mathematical expression is:

[0089]

[0090] in, The table is based on the baseline health and safety factor constructed from the actual operating status of the motor, which is used to reflect the basic operating health level of the motor under the current operating conditions; The abnormal deviation coefficient represents the degree of deviation between the actual operating state and the predicted state of the digital twin model, and is used to reflect the degree of disruption of the consistency between the virtual and the real. , , which is a weighting coefficient used to adjust the contribution ratio of the baseline health assessment and the abnormal deviation assessment to the overall health assessment result under different working conditions.

[0091] In this embodiment, the weighting coefficient , It can adaptively adjust to the operating conditions such as motor speed, load rate, and cooling conditions. One way to achieve this adaptive adjustment is as follows:

[0092]

[0093]

[0094] in, For real-time load rate, For real-time rotational speed, For the rated speed, For real-time intake air temperature, The maximum allowable inlet air temperature is specified in the design. , , , The adjustment coefficient is calibrated according to the specific motor model and application scenario, and its value does not constitute a limitation on the present invention.

[0095] The aforementioned adaptive weight adjustment mechanism embodies the following engineering logic: Under high load, high speed, or deteriorating cooling conditions, the reliability of threshold-based baseline health assessments decreases relatively, while the anomaly sensitivity based on virtual-to-real deviations increases, thereby increasing the reliability of the assessment. Strengthen the ability to identify potential anomalies.

[0096] In this embodiment, the benchmark safety factor The mathematical expression is,

[0097]

[0098] in, , , The three parts correspond to the temperature safety factor, vibration safety factor, and bearing temperature safety factor, respectively, and satisfy the following conditions: The weighting coefficients are set to reflect the degree of influence of different monitoring quantities on the overall health assessment, and can be adjusted according to the motor structure and application scenario.

[0099] In this embodiment, considering the sensitivity of permanent magnet synchronous motors to temperature changes, for example, 80°C is used as the reference temperature for the motor, and 130°C is used as the insulation tolerance margin, with a weighting of 1 / 30°C. , , This value is only an example of implementation and does not constitute a limitation of the present invention.

[0100] In this embodiment, the abnormal deviation coefficient The mathematical expression is:

[0101]

[0102] in, The residual value represents the difference between the measured temperature of the permanent magnet and the predicted value of the digital twin model. It is used to reflect potential demagnetization or local thermal anomalies. Theoretically, the two should be consistent. If the residual value increases significantly, it indicates that local demagnetization of the permanent magnet or deterioration of heat dissipation may occur. The residual between the measured and predicted values ​​of the vibration characteristic quantity of the casing is used to indicate whether a significant vibration problem has occurred. This represents the residual between the measured and predicted cooling airflow speeds, used to reflect the level of consistency between the cooling system's operating status and the virtual prediction. , , These are the standard deviations of each residual obtained from the historical data of the motor's normal operation, used to normalize residuals of different dimensions. In this embodiment, these standard deviations can be obtained statistically based on historical data from the initial stable operation phase. =3.0℃, =0.08mm / s, =0.05A, the value of which is given as an example and does not constitute a limitation of the present invention.

[0103] This embodiment also provides a health assessment and early warning system for permanent magnet synchronous motors based on digital twins. This system is used to implement the aforementioned health assessment and early warning method for permanent magnet synchronous motors based on digital twins. Figure 6 As shown, the system includes a data acquisition module, a virtual model module, a data analysis and processing module, and a visualization health assessment and early warning module.

[0104] The data acquisition module is used to collect multi-parameter operating data of the target motor in real time, and to preprocess the multi-parameter data to build a multi-source entity operating database that is dynamically updated over time. This database is used to characterize the actual operating status of the motor under different operating conditions and to provide a data basis for subsequent health assessment.

[0105] The virtual model module is used to construct a refined 3D model and a multi-physics coupled digital twin model of the target permanent magnet synchronous motor. Based on the model order reduction method, it realizes the online operation capability of the virtual model to obtain virtual predictive operation data consistent with the actual operation mechanism of the motor and constructs a virtual model operation database.

[0106] The data analysis and processing module is used to integrate the virtual model operation data and the motor physical operation data. Based on the virtual-real consistency evaluation mechanism, it quantitatively evaluates the motor's operating health status, calculates the comprehensive health index under the current working condition by constructing a comprehensive safety factor model, and automatically determines the health level according to the relationship between the comprehensive health index and the preset threshold range. The evaluation results are then transmitted to the visualization health assessment and early warning module to trigger the corresponding early warning response.

[0107] The visualization health assessment and early warning module is used to dynamically visualize the operating status of the motor entity, the operating status of the digital twin model, and the comprehensive health assessment results. It also provides real-time early warning display and handling guidance based on the assessment results from the data analysis and processing module according to the hierarchical health early warning response rules. At the same time, the module is also used to collect operation feedback signals during human-computer interaction and feed these feedback signals back to the data analysis and processing module to support the closed-loop optimization of model parameters and health assessment strategies.

[0108] Furthermore, the data acquisition module includes a sensor submodule, a signal processing submodule, and a communication submodule. The sensor submodule is a sensor array composed of multiple types of data sensors arranged in a predetermined pattern, used to collect multi-source information such as electrical, thermal, and vibration data during motor operation. The signal processing submodule performs signal conditioning, analog-to-digital conversion, data cleaning, feature extraction, and data standardization via a host computer. The communication submodule transmits the acquired and preprocessed data to the data analysis and processing module to support the online operation and health assessment process of the digital twin model.

[0109] Furthermore, the virtual model module includes a 3D modeling submodule, a numerical simulation submodule, and a partitioned model reduction submodule. The numerical simulation submodule is used to generate a multiphysics simulation dataset under multiple operating conditions, and inputs the simulation data into the partitioned model reduction submodule for processing to obtain a dynamic virtual model runtime database that meets real-time operation requirements.

[0110] Furthermore, the data analysis and processing module can be implemented through a program and integrated into a host computer or edge computing unit to perform functions such as virtual and real data fusion, comprehensive health indicator calculation, and health level determination.

[0111] Furthermore, the visualized health assessment and early warning module can be implemented by combining a 3D visualization platform with a custom program. Its feedback signal feedback module can generate structured feedback signals through interface interaction buttons and send the feedback signals back to the corresponding interface of the data analysis and processing module to trigger the online learning and optimization process of the digital twin model and health assessment strategy.

[0112] Furthermore, based on the calculation of comprehensive health indicators and the determination of health level by the data analysis and processing module, the operating status of the target permanent magnet synchronous motor can be warned and determined according to the preset hierarchical health warning response rules. In this embodiment, the hierarchical health warning is described in a three-level manner, but the specific number of levels for health warning is not limited.

[0113] In this embodiment, when When the value is ≥0.85, it indicates that the target permanent magnet synchronous motor is in a relatively good health state, the system judges it to be in normal operation, and the visual health assessment and early warning interface displays green; when 0.70≤ When the value is less than 0.85, it indicates a slight decline in the motor's operational health. The system classifies this as a "priority" health status, displayed in yellow on the interface, prompting operators to continuously monitor and inspect the motor's operating status. When 0.50 ≤ When the value is less than 0.70, it indicates a significant abnormality in the motor's operational health. The system determines this as a maintenance-level health warning state, displays an orange color on the interface, and flashes a warning, sending corresponding health intervention and maintenance suggestions to maintenance personnel. When the value is less than 0.50, it indicates that the motor's operating health condition has seriously deteriorated. The system determines that it is in a state of serious health risk, triggers the local audible and visual alarm device, and links the control system to execute emergency protection strategies. At the same time, it automatically saves key operating data before and after the fault and generates health degradation and fault analysis data files.

[0114] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0116] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0118] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for health assessment and early warning of permanent magnet synchronous motors based on digital twins, characterized in that, Includes the following steps: Step S1: Collect the operating parameters of the permanent magnet synchronous motor in real time, preprocess the operating parameters, and construct a multi-source entity operation database that is dynamically updated over time. Step S2: Construct a digital twin model that is consistent with the actual operating mechanism of the permanent magnet synchronous motor, and obtain virtual predicted operating data corresponding to the operating parameters; Step S3: Perform model reduction processing on the digital twin model to transform the high-dimensional multiphysics coupling model into a reduced-order virtual model that can run online, so as to meet the real-time computational requirements of real-time health assessment and early warning analysis. Step S4: Based on the virtual prediction operation data of the multi-source entity operation database and the reduced-order virtual model, perform virtual-real consistency assessment and calculate a comprehensive health index, wherein the comprehensive health index is obtained by fusing the baseline health assessment result based on the entity operation status and the abnormal deviation assessment result based on the degree of virtual-real deviation. Step S5: Compare the comprehensive health index with the preset threshold range to determine the motor health level, and trigger the corresponding graded health warning response according to the motor health level.

2. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, In step S1, the operating parameters include at least two or more of the following: current, voltage, speed, temperature, and vibration parameters, and further include one or more of the following cooling-related operating parameters: inlet air temperature, air velocity, air volume, and heat exchange boundary characterization parameters.

3. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, In step S1, the preprocessing includes: data cleaning, missing value handling, outlier removal, time alignment, format unification, standardization processing, and storage index management. The preprocessing process includes: Step S101: Perform preliminary screening on the raw multi-parameter datasets from different data sources, removing erroneous values, missing values, and outlier values; Step S102: Convert the cleaned multi-parameter data into a unified data structure and format and complete time alignment; Step S103: Perform encoding and compression processing on the standardized data; Step S104: Write the encoded and compressed multi-parameter data into the multi-source entity runtime database; Step S105: Classify and organize the data and create an index in the multi-source entity running database.

4. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, Step S2 specifically includes: An electromagnetic solution model is established based on the motor structure and material parameters to obtain electromagnetic force and loss. The temperature distribution is obtained by inputting the loss as a heat source into the thermal field model and combining it with heat transfer boundary conditions. Electromagnetic force is used as the excitation input to obtain the vibration response of the structural dynamics model, and the equivalent correlation between the vibration response and the radiated noise characteristics is further established to form a digital twin model with multi-physics coupling. The digital twin model includes at least two or more of the following: electromagnetic field model, thermal field model, and structural dynamics field model, and is used to reflect the multi-physics field coupling operation mechanism inside the motor.

5. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, Step S3 employs a partitioned model order reduction strategy, reducing the order of different physical field sub-models separately and integrating them at the system level, including: Acquire multi-condition simulation snapshot data; Feature extraction and dimensionality reduction are performed on temperature field snapshot data. A fast prediction model for the temperature field is constructed with operating parameters as input and low-dimensional features as output. A low-order model of the structural dynamic response is established, and an equivalent transfer relationship between vibration response and noise characteristics is constructed. Configure data interfaces for each reduced-order sub-model and integrate them into a system-level real-time reduced-order virtual model.

6. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, In step S4, the virtual-real consistency assessment includes: calculating the deviation or residual between the measured parameters of at least one type of entity and the virtual predicted parameters, and using the deviation or residual as the input of the abnormal deviation assessment result to reflect the degree of deviation of the motor operating state from the mechanism consistency prediction result. The fusion weight of the baseline health assessment result and the abnormal deviation assessment result in the comprehensive health index is adaptively adjusted according to the operating conditions. The operating conditions include at least one or more of speed, load rate and cooling conditions, so as to improve the sensitivity to the abnormal deviation assessment results when under high load, high speed or deteriorating cooling conditions.

7. The method for health assessment and early warning of permanent magnet synchronous motors based on digital twins according to claim 1, characterized in that, In step S5, the motor health level is classified into at least three levels. The corresponding motor health level is output based on the comparison results of the comprehensive health index and at least two preset threshold intervals, and corresponding early warning prompts and handling guidelines are configured for different motor health levels.

8. A health assessment and early warning system for permanent magnet synchronous motors based on digital twins, used to implement the method described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to collect motor operating parameters and build a multi-source entity operating database; The virtual model module is used to construct a digital twin model that is consistent with the actual operating mechanism of the motor and output virtual predictive operating data. It also includes a reduction unit for model reduction processing to form a reduced virtual model that can run online. The data analysis and processing module is used to integrate the multi-source entity operation database and the virtual prediction operation data, perform virtual-real consistency assessment and calculate comprehensive health indicators, and determine the motor health level based on the comprehensive health indicators; The visualization health assessment and early warning module is used to display the physical operating status, virtual prediction status and health assessment results, and trigger graded health early warning responses based on the motor health level.

9. The health assessment and early warning system for permanent magnet synchronous motors based on digital twins according to claim 8, characterized in that, The data analysis and processing module is configured to evaluate the consistency between the actual motor operation data and the predicted data of the digital twin model, and generate anomaly deviation evaluation results based on the degree of deviation between the actual operation state and the virtual predicted state. The data analysis and processing module is further configured to integrate the baseline health assessment results based on the entity's operating status with the abnormal deviation assessment results to generate a comprehensive health index that characterizes the overall operating health level of the motor.

10. The health assessment and early warning system for permanent magnet synchronous motors based on digital twins according to claim 8, characterized in that, The visual health assessment and early warning module is configured to classify the motor operating status and output corresponding health early warning information based on the comparison results of the comprehensive health indicators and the preset threshold range. The visualized health assessment and early warning module is also configured to collect human-computer interaction feedback information and send the feedback information back to the data analysis and processing module for adaptive optimization of model parameters or health assessment strategies.

Images