Motor model construction method and device

Abstract

The invention provides a motor model construction method and device. The method comprises the steps of obtaining motor physical parameters of a target traction motor; constructing a corresponding finite element analysis model according to the physical parameters of the motor; performing mesh generation on the finite element analysis model according to the heat transfer parameters of each component in the finite element analysis model to obtain a heat transfer model; performing heat transfer simulation according to the heat transfer model to obtain corresponding thermal characteristics; obtaining a first traction motor model based on the finite element analysis model and the corresponding thermal characteristics; and according to a preset operation condition threshold value of the target traction motor, performing model order reduction on the first traction motor model to obtain a second traction motor model.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and apparatus for constructing an electric motor model. Background Technology

[0002] A traction motor is an electric motor used to drive vehicles. It converts electrical energy into mechanical energy to provide power to the vehicle. The traction motor is the core component of an electric traction system, and its performance directly affects the vehicle's operating efficiency and reliability. When abnormal conditions occur during the operation of a traction motor, these abnormalities may lead to a decrease in motor performance, reduced efficiency, or even complete cessation of operation.

[0003] Currently, traction motor fault diagnosis relies on the precise analytical solution of model indicators to construct the residual between the analytical solution and the actual system output indicators. Signal analysis techniques are then used to extract the traction motor's operational characteristics, and fault diagnosis is performed based on residual evaluation according to signal change trends. While this mathematical model-based approach is suitable for linear systems, it requires simplification when dealing with complex, nonlinear, strongly coupled, and multivariable systems, leading to an inability to accurately describe the overall motor characteristics. Summary of the Invention

[0004] This application provides a method and apparatus for constructing a motor model, which can accurately and comprehensively describe the characteristics of the entire motor.

[0005] In a first aspect, this application provides a method for constructing a motor model, the method comprising:

[0006] Obtain the physical parameters of the target traction motor;

[0007] Based on the physical parameters of the motor, a corresponding finite element analysis model is constructed;

[0008] According to the heat transfer parameters of each component in the finite element analysis model, the finite element analysis model is meshed to obtain the heat transfer model;

[0009] According to the heat transfer model, heat transfer simulation is performed to obtain the corresponding thermal characteristics;

[0010] Based on the finite element analysis model and the corresponding thermal characteristics, the first traction motor model is obtained;

[0011] Based on the preset operating condition threshold of the target traction motor, the first traction motor model is reduced in order to obtain the second traction motor model.

[0012] In some possible implementations, after reducing the order of the first traction motor model according to the preset operating condition threshold of the target traction motor to obtain the second traction motor model, the method further includes:

[0013] Based on the operating status of the target traction motor, the actual operating data of the target traction motor is obtained;

[0014] Based on a preset parameter identification algorithm, the actual operating data is subjected to parameter identification to determine the key parameters in the actual operating data.

[0015] Based on the key parameters, the corresponding correction values ​​are obtained;

[0016] The second traction motor model is modified according to the aforementioned correction value.

[0017] In some possible implementations, obtaining the corresponding correction value based on the key parameter includes:

[0018] Based on the first traction motor model, simulations are performed on the key parameters to obtain simulation results corresponding to the key parameters;

[0019] The simulation results are compared with the key parameters to obtain the simulation error;

[0020] Based on the preset parameter identification algorithm and the simulation error, the corresponding correction value is obtained.

[0021] In some possible implementations, modifying the second traction motor model according to the modification value includes:

[0022] Replace the corresponding parameter values ​​in the second traction motor model according to the correction values;

[0023] Once the parameter values ​​of the second traction motor model have been replaced, the second traction motor model is recalibrated.

[0024] In some possible implementations, obtaining the actual operating data of the target traction motor based on its operating state includes:

[0025] Collect relevant data on the operating status of the target traction motor;

[0026] The noise reduction preprocessing of the operating status-related data is performed to obtain the actual operating data of the target traction motor.

[0027] In some possible implementations, the target traction motor's physical parameters include at least one of electrical parameters, geometric parameters, and material parameters.

[0028] In some possible implementations, the step of reducing the order of the first traction motor model to obtain a second traction motor model according to the preset operating condition threshold of the target traction motor includes:

[0029] Obtain the preset electrical operating condition threshold and thermal operating condition threshold of the target traction motor;

[0030] Based on the electrical characteristics of the first traction motor model, the electromagnetic model of the first traction motor model is reduced in order using equivalent circuit extraction.

[0031] Based on the thermal characteristics of the first traction motor model, the heat transfer model of the first traction motor model is reduced in order using a reduced-order model to obtain the second traction motor model.

[0032] In some possible implementations, after reducing the order of the first traction motor model according to the preset operating condition threshold of the target traction motor to obtain the second traction motor model, the method further includes:

[0033] The preset fault conditions are input into the second traction motor model, and simulation experiments are conducted through the second traction motor model to obtain simulation results;

[0034] Extract the reference feature signal corresponding to the preset fault condition from the simulation results.

[0035] In some possible implementations, after extracting the feature parameter signals from the simulation results, the method further includes:

[0036] The operating conditions of the traction motor to be tested are input into the second traction motor model to obtain the parameter signals of the traction motor to be tested;

[0037] The parameter signal of the traction motor to be tested is matched with the reference feature signal to obtain the matching result;

[0038] Based on the matching results, the fault condition of the traction motor to be tested is determined.

[0039] In some possible implementations, extracting the reference feature signal corresponding to the preset fault condition from the simulation results includes:

[0040] Based on the differences between the stable operating state and the faulty operating state in the simulation results, the corresponding signal residuals are obtained;

[0041] Extract the time-frequency domain changes from the feature signals in the simulation results to obtain the time-frequency domain change signal;

[0042] The spatial variation signal is obtained by extracting the spatial variation from the feature signals in the simulation results;

[0043] Based on the signal residual, the time-frequency domain variation signal, and the spatial variation signal, a reference characteristic signal corresponding to the preset fault condition is obtained.

[0044] Secondly, this application provides a motor model building device, the device comprising:

[0045] The acquisition module is used to acquire the physical parameters of the target traction motor.

[0046] The construction module is used to construct the corresponding finite element analysis model based on the physical parameters of the motor;

[0047] The meshing module is used to mesh the finite element analysis model according to the heat transfer parameters of each component in the finite element analysis model, so as to obtain a heat transfer model.

[0048] The simulation module is used to perform heat transfer simulation according to the heat transfer model to obtain the corresponding thermal characteristics;

[0049] The determination module is used to obtain the first traction motor model based on the finite element analysis model and the corresponding thermal characteristics;

[0050] The model reduction module is used to reduce the model order of the first traction motor model according to the preset operating condition threshold of the target traction motor, so as to obtain the second traction motor model.

[0051] The motor model construction method and apparatus provided in this application embodiment obtain the physical parameters of the target traction motor, construct a corresponding finite element analysis model based on the physical parameters, mesh the finite element analysis model according to the heat transfer parameters of each part in the finite element analysis model to obtain a heat transfer model, and then perform heat transfer simulation according to the heat transfer model to obtain the corresponding thermal characteristics. Based on the finite element analysis model and the corresponding thermal characteristics, a first traction motor model is obtained. Then, according to the operating condition threshold, the order of the first traction motor model is reduced to obtain a second traction motor model. In this way, this application embodiment simulates the behavior of the motor under various operating conditions, comprehensively considers the physical characteristics of the motor in different aspects, and constructs a high-precision digital twin model, thereby accurately and comprehensively describing the characteristics of the entire motor. Attached Figure Description

[0052] This application can be better understood from the following description of specific embodiments in conjunction with the accompanying drawings, wherein:

[0053] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings, wherein the same or similar reference numerals denote the same or similar features.

[0054] Figure 1 This is a flowchart of a motor model construction method provided in one embodiment of this application;

[0055] Figure 2 This is a flowchart of a motor model construction method provided in another embodiment of this application;

[0056] Figure 3 This is a schematic flowchart of motor fault diagnosis provided in one embodiment of this application;

[0057] Figure 4 This is a flowchart of the correction process for a digital twin model of a traction motor provided in one embodiment of this application;

[0058] Figure 5 This is a schematic diagram of the three-phase current curves of the motor during a phase loss fault;

[0059] Figure 6 This is a schematic diagram comparing the non-open-circuit current of a motor with a phase loss fault and a normal motor.

[0060] Figure 7 This is a schematic diagram of the changes in the three-phase current of the motor during a stall fault;

[0061] Figure 8 This is a schematic diagram of the stator copper loss variation curve during a stall fault;

[0062] Figure 9 This is a schematic diagram of the current curves of the three phases of the motor before and after the coupling completely breaks;

[0063] Figure 10 This is a schematic diagram of the motor's thermal field before and after the coupling completely breaks;

[0064] Figure 11 This is a schematic diagram of the structure of a motor model building device provided in one embodiment of this application;

[0065] Figure 12 This is a schematic diagram of the hardware structure of the motor model building device provided in the embodiments of this application. Detailed Implementation

[0066] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0067] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0068] To address the problems of the prior art, embodiments of this application provide a method, apparatus, device, storage medium, and product for constructing a motor model. The method for constructing a motor model provided in this application embodiment will be described first below.

[0069] Figure 1 A flowchart illustrating a method for constructing a motor model according to an embodiment of this application is shown. Figure 1 As shown, the method includes the following steps: S101 to S105.

[0070] S101: Obtain the physical parameters of the target traction motor.

[0071] In practical implementation, the physical parameters of the motor can include: the motor's geometric parameters, such as the dimensions, shape, and materials of the stator, rotor, core, and windings; material properties, such as the electrical conductivity, thermal conductivity, density, and specific heat capacity of various parts of the motor; and the motor's operating conditions, such as current, voltage, frequency, and speed. As an example of S101, the geometric dimensions of the motor can be imported from the physical modeling tool, and the physical properties of different materials can be obtained by calling the material library.

[0072] As another example of S101, relevant data such as the temperature rise and efficiency of the motor can be obtained through experiments, and physical parameters can be obtained through curve fitting.

[0073] S102: Based on the above physical parameters of the motor, construct the corresponding finite element analysis model.

[0074] In the specific implementation, based on the geometric parameters obtained from S101, a three-dimensional geometric model of the motor is created in the finite element analysis software. An appropriate mesh type is selected, the mesh density is set according to the complexity of the model, and each mesh element is assigned corresponding material properties, such as electrical conductivity and thermal conductivity.

[0075] S103: Based on the heat transfer parameters of each component in the above finite element analysis model, mesh the above finite element analysis model to obtain the heat transfer model.

[0076] In practical implementation, the finite element model of the motor includes not only electromagnetic and mechanical parameters, but also heat transfer parameters such as thermal conductivity, temperature boundary conditions, and ambient temperature. For areas involving heat transfer, such as windings, stator cores, and rotors, more detailed meshing can be performed because heat conduction typically requires high precision. This ensures that each element can effectively transfer heat, including setting thermal boundaries and temperature sources in adjacent areas, resulting in a heat transfer model.

[0077] S104: Perform heat transfer simulation according to the above heat transfer model to obtain the corresponding thermal characteristics.

[0078] In practical implementation, the temperature distribution inside the motor is numerically solved using heat conduction equations, such as steady-state or transient heat conduction equations, based on the model and boundary conditions. External cooling conditions, such as air cooling, liquid cooling, and natural convection, as well as internal heat sources, such as current losses, are set to calculate and output the corresponding thermal characteristics, such as the temperature field and heat flux density, of different motor components.

[0079] S105: Based on the above finite element analysis model and the corresponding thermal characteristics, the first traction motor model is obtained.

[0080] In practical implementation, the electromagnetic fields of the motor, such as current distribution and magnetic field, are coupled with the thermal field. Based on the coupled model, the performance of the motor under different loads, speeds, and operating temperatures is calculated, such as efficiency, power loss, and temperature rise, to obtain the first traction motor model.

[0081] S106: Based on the preset operating condition threshold of the target traction motor, the first traction motor model is reduced in order to obtain the second traction motor model.

[0082] In the specific implementation, reduction methods such as principal component analysis, singular value decomposition, and Latin hypercube sampling are used to reduce the high-dimensional finite element model to a simplified low-dimensional model. Operating thresholds are set based on the motor's operating conditions, such as maximum temperature and maximum power loss. Based on these thresholds, unimportant physical processes or regions are removed from the model, retaining only key influencing factors. From the reduced-dimensional model, a model that can run with lower computational resources is generated. This model, while meeting accuracy requirements, can quickly respond to the motor's dynamic behavior, resulting in the second traction motor model.

[0083] The motor model construction method provided in this application involves obtaining the physical parameters of the target traction motor, constructing a corresponding finite element analysis model based on these parameters, meshing the finite element analysis model according to the heat transfer parameters of each part, obtaining a heat transfer model, simulating heat transfer according to the heat transfer model, obtaining the corresponding thermal characteristics, obtaining a first traction motor model based on the finite element analysis model and the corresponding thermal characteristics, and then reducing the order of the first traction motor model according to the operating condition threshold to obtain a second traction motor model. By simulating the behavior of the motor under various operating conditions and comprehensively considering the physical characteristics of the motor in different aspects, a high-precision digital twin model is constructed, thereby accurately and comprehensively describing the characteristics of the entire motor.

[0084] In order to maintain high accuracy under various complex working conditions, in some embodiments, after S106, the method may further include the following steps: S107 to S110.

[0085] S107: Obtain the actual operating data of the target traction motor based on its operating status.

[0086] In practical implementation, various sensors can be deployed on key components of the motor, such as the stator, rotor, windings, and bearings. These sensors can include temperature sensors, current sensors, speed sensors, vibration sensors, and torque sensors. These sensors record the motor's parameters in real time during operation. Real-time data acquisition, such as through industrial control systems or dedicated data acquisition cards, collects the signals output by these sensors. Monitoring software then records the motor's operating status, such as speed, temperature, and current, in real time. By analyzing the motor's current and voltage waveforms, its operating status is assessed. Temperature data from various parts of the motor is obtained through temperature sensors, and the mechanical output characteristics of the motor are acquired through speed and torque sensors to determine the motor's operating load. Combining this information, the actual operating data of the target traction motor is obtained.

[0087] S108: Based on the preset parameter identification algorithm, perform parameter identification on the above actual operating data to determine the key parameters in the above actual operating data.

[0088] In practical implementation, key parameters are extracted from actual operating data using regression analysis methods such as least squares and ridge regression. For example, motor losses and efficiency can be calculated using parameters such as input voltage, current, speed, and output power. Optimization algorithms such as particle swarm optimization, genetic algorithms, and simulated annealing are used to adjust the parameters in the model based on actual operating data and the objective function. System identification methods such as Kalman filtering, recursive least squares, and minimum mean square error algorithms are used to dynamically adjust the motor model parameters based on actual operating data, and parameter estimation is performed on the dynamic motor model using input and output data. Electromagnetic losses, mechanical losses, and iron losses of the motor are calculated using current, voltage, and speed data, and their variation patterns under different operating conditions are determined. Motor efficiency is calculated using actual operating data and theoretical values ​​to determine which factors have the greatest impact on efficiency. Combined with temperature sensor data, temperature changes of various motor components are analyzed to identify thermal parameters that may affect motor performance, thereby determining the key parameters in the aforementioned actual operating data.

[0089] S109: Based on the above key parameters, obtain the corresponding correction values.

[0090] In practical implementation, if the identification algorithm determines that the motor loss is greater than the model's predicted loss, the model can be corrected by increasing the motor's loss coefficient, adjusting winding resistance, iron loss, and other parameters. Based on temperature data and thermal analysis results, parameters such as heat conduction, heat convection, and heat radiation in the model are adjusted to ensure the model can accurately predict the motor's temperature field. If actual operating data indicates that the motor's efficiency is lower than expected, efficiency-related parameters in the model, such as rotor resistance, stator current, and magnetic field saturation, can be adjusted to better reflect actual performance. For each key parameter, the system calculates a correction coefficient. For example, for losses, it may be necessary to adjust the permeability of the motor core material or adjust the winding resistance. The final correction value is obtained, which can be an incremental value, such as the deviation in temperature or efficiency, or a proportional factor, such as the ratio of motor torque to current.

[0091] S110: Correct the second traction motor model as described above according to the correction values.

[0092] In the specific implementation, the software modifies key parameters in the motor model, such as material properties, current density, loss coefficient, and thermal conductivity, and applies the corrected values ​​to the second traction motor model. The updated model needs to be re-coupled and calculated in electromagnetic, thermal, and mechanical fields, and the electromagnetic effects, temperature field, and mechanical response of the motor are recalculated based on the new corrected values. The corrected model is then re-simulated to determine the corrected second traction motor model.

[0093] The above-described implementation method of this application obtains the actual operating data of the target traction motor based on its operating state. Then, it identifies the parameters of the actual operating data according to a preset parameter identification algorithm, determines the key parameters in the actual operating data, and obtains the corresponding correction values ​​based on the key parameters. The second traction motor model is then corrected according to the correction values, which ensures that the motor model always reflects its true working state. The feedback-based data-driven method enables the motor model to maintain high accuracy under various complex working conditions.

[0094] To improve simulation accuracy, in some implementations, the above-mentioned S109 may include the following steps: S1091 to S1093.

[0095] S1091: Based on the first traction motor model mentioned above, simulations are performed on the key parameters mentioned above to obtain simulation results corresponding to the key parameters mentioned above.

[0096] In the specific implementation, a motor model is built using a motor simulation software platform, and the previously established first traction motor model is imported into the simulation environment. This model should include the motor's electrical, mechanical, and thermal parameters. Simulation conditions are set, the simulation model is run, and the motor's behavior under the set conditions is calculated using a step-time algorithm to obtain simulation results corresponding to the aforementioned key parameters.

[0097] S1092: Compare the above simulation results with the above key parameters to obtain the simulation error.

[0098] In the specific implementation, key parameters such as motor temperature, efficiency, current, voltage, speed, and torque are extracted from the actual operating data obtained in the previous steps. Error indices are defined, and the simulation results are compared with the actual data item by item to calculate the error value of each parameter. For example, by comparing the simulation results with the actually measured indicators such as temperature, efficiency, and losses, the errors of each indicator are obtained, and the simulation error is determined.

[0099] S1093: Based on the preset parameter identification algorithm and the above simulation error, obtain the corresponding correction value.

[0100] In the specific implementation, based on the magnitude and type of the error, an appropriate parameter identification algorithm is selected to adjust the key parameters in the model. Based on the error information, the correction parameters are calculated to obtain the corresponding correction values.

[0101] The above-described implementation method of this application simulates the key parameters based on the first traction motor model to obtain simulation results corresponding to the key parameters. Then, the simulation results are compared with the key parameters to obtain the simulation error. Subsequently, a corresponding correction value is obtained according to the preset parameter identification algorithm and the simulation error, ensuring that the motor model can gradually approach the working state of the actual motor through continuous correction, thereby improving the simulation accuracy.

[0102] To ensure the reliability and accuracy of the model under different operating conditions, in some implementations, the above-mentioned S110 may include the following steps: S1101 to S1102.

[0103] S1101: Replace the corresponding parameter values ​​in the second traction motor model above with the above correction values.

[0104] In the actual implementation, the correction values ​​are obtained, the second traction motor model is loaded, and it is determined which parameters need to be updated based on the correction values. Typically, the correction values ​​correspond one-to-one with various physical characteristics of the motor model, such as electrical parameters, thermal parameters, and mechanical parameters. In the second traction motor model, the corresponding parameters are replaced according to the correction values.

[0105] S1102: After the parameter values ​​of the second traction motor model have been replaced, the second traction motor model is recalibrated.

[0106] In practice, after the parameter values ​​of the second traction motor model are replaced, the calibration purpose and method are confirmed, and the second traction motor model is recalibrated to ensure that the updated model can simulate the behavior of the motor in actual operation as accurately as possible.

[0107] The above-described implementation method of this application replaces the corresponding parameter values ​​in the second traction motor model according to the above-described correction values. After the parameter value replacement of the second traction motor model is completed, the second traction motor model is recalibrated to ensure that the motor model can accurately reflect the behavior of the actual motor under the new parameters, and to ensure the reliability and accuracy of the model under different working conditions.

[0108] In order to ensure that the collected signals can accurately reflect the actual operating conditions of the motor, in some embodiments, the above-mentioned S107 may include the following steps: S1071 to S1072.

[0109] S1071: Collect relevant data on the operating status of the target traction motor.

[0110] In the specific implementation, the key operating parameters of the target motor are clearly defined, such as current, voltage, speed, temperature, power, torque, and efficiency. Based on the operating conditions of the target traction motor, an appropriate sampling frequency and data accuracy are selected. Then, the collected data is monitored in real time to ensure the stability of the data stream, thereby collecting relevant data on the operating status of the target traction motor.

[0111] S1072: Perform noise reduction preprocessing on the above-mentioned operating status data to obtain the actual operating data of the target traction motor.

[0112] In the specific implementation, appropriate noise reduction methods are selected, such as filtering, wavelet transform and data smoothing, and the type and pattern of noise are identified. The above-mentioned operating status related data are subjected to noise reduction preprocessing to obtain the actual operating data of the target traction motor.

[0113] The above-described embodiments of this application collect relevant data on the operating status of the target traction motor, and then perform noise reduction preprocessing on the aforementioned relevant data to obtain the actual operating data of the target traction motor. By using a variety of data processing techniques, high-quality input data is provided for subsequent analysis and modeling, ensuring that the collected signals can truly reflect the actual operating conditions of the motor.

[0114] In some embodiments, the physical parameters of the target traction motor include at least one of electrical parameters, geometric parameters, and material parameters.

[0115] In order to improve the efficiency of model calculation, in some implementations, the above S106 includes the following steps: S1061 to S1063.

[0116] S1061: Obtain the preset electrical operating condition threshold and thermal operating condition threshold of the target traction motor.

[0117] In practice, electrical operating condition thresholds are obtained through actual operating data and operating condition monitoring equipment. These thresholds typically include the motor's rated voltage, current, and power range. Thermal operating thresholds mainly involve the motor's temperature rise control, including stator temperature, rotor temperature, and winding temperature. Based on this information, the preset electrical and thermal operating condition thresholds for the target traction motor are determined.

[0118] S1062: Based on the electrical characteristics of the first traction motor model described above, the electromagnetic model of the first traction motor model is reduced in order using equivalent circuit extraction.

[0119] In the specific implementation, based on the electrical characteristics of the first traction motor model mentioned above, the redundant electromagnetic model in the motor is simplified into a smaller circuit network through equivalent circuit simplification, and a high-order, complex electromagnetic model is simplified into a low-order model. These order reduction operations can be performed using linear system analysis tools.

[0120] S1063: Based on the thermal characteristics of the first traction motor model described above, the heat transfer model of the first traction motor model is reduced in order using a reduced-order model to obtain the second traction motor model.

[0121] In the specific implementation, based on the thermal characteristics of the first traction motor model mentioned above, a heat flow-based reduced-order model is used. For example, the thermal system is simplified from a continuous three-dimensional heat conduction model to a one-dimensional or two-dimensional model, or the heat transfer equation is simplified by the relationship between thermocouples and temperature distribution. Finally, the reduced-order electromagnetic model and the reduced-order thermal model are combined to obtain the second traction motor model.

[0122] The above-described implementation method of this application obtains the preset electrical operating condition threshold and thermal operating condition threshold of the target traction motor, and then, based on the electrical characteristics of the first traction motor model, uses equivalent circuit extraction to reduce the order of the electromagnetic model of the first traction motor model. Furthermore, based on the thermal characteristics of the first traction motor model, uses a reduced-order model to reduce the order of the heat transfer model of the first traction motor model to obtain the second traction motor model. By simplifying the complexity of the model, the model calculation efficiency and real-time analysis capability are improved.

[0123] In order to extract data that reflects the characteristics of the fault, in some implementations, reference is made to... Figure 2 After S106 above, the method may further include the following steps: S201 to S202.

[0124] S201: Input the preset fault conditions into the second traction motor model mentioned above, and conduct a simulation experiment through the second traction motor model to obtain the simulation results.

[0125] In the specific implementation, some preset fault conditions are defined in advance. These fault conditions are usually types of faults that may affect the performance of the traction motor, obtained based on engineering experience or through historical data analysis. Examples include motor winding short circuits, overloads, excessive temperature rise, rotor imbalance, and other fault conditions. The parameters of these preset fault conditions are converted into an input format that the model can understand. These inputs are then passed to the second traction motor model, and simulation calculations are performed based on the inputs to obtain the dynamic response results of the motor under these fault conditions, thus yielding the simulation results.

[0126] S202: Extract the reference feature signal corresponding to the preset fault condition from the above simulation results.

[0127] In the specific implementation, after obtaining the simulation results, feature signals that are helpful for fault analysis are extracted from the simulation results. Relevant signals are selected for extraction based on the fault type. For example, if the focus is on electrical faults in the motor, the current waveform might be the primary focus; if it is a mechanical fault, the speed or vibration signal might be the primary focus. Signal preprocessing is then performed, including denoising, smoothing, and normalization. Because the simulation data may contain noise, signal preprocessing is necessary to make the extracted features more representative and reliable. Finally, key features are extracted from the preprocessed signals to obtain the reference feature signals corresponding to the preset fault conditions.

[0128] The above-described implementation method of this application involves inputting a preset fault condition into the second traction motor model, conducting a simulation experiment using the second traction motor model, obtaining simulation results, and then extracting the reference feature signal corresponding to the preset fault condition from the simulation results. The simulation is used to simulate the behavior of the motor under fault conditions, and data that can reflect the fault characteristics are extracted from the simulation results.

[0129] In order to achieve automatic diagnosis of traction motor faults, in some embodiments, after S202, the method may further include the following steps: S203 to S205.

[0130] S203: Input the operating conditions of the traction motor to be tested into the second traction motor model mentioned above to obtain the parameter signals of the traction motor to be tested.

[0131] In the specific implementation, the operating conditions to be tested are input into the second traction motor model. After these operating condition data are input, the second traction motor model will perform simulation calculations to simulate the dynamic behavior of the motor under these conditions and obtain the parameter signals of the traction motor to be tested.

[0132] S204: Match the parameter signal of the traction motor to be tested with the reference feature signal to obtain the matching result.

[0133] In the specific implementation, the real-time parameter signals of the motor under test are compared with the reference feature signals obtained from previous simulations. During the matching process, the software compares the similarity between the signal under test and each reference signal, and outputs a matching result. These matching results represent the probability score of the motor under test under different fault conditions.

[0134] S205: Based on the above matching results, determine the fault condition of the above-mentioned traction motor to be tested.

[0135] In practice, based on the matching results in step S204, the fault type and severity of the motor to be tested are determined, and the fault status of the traction motor to be tested is identified.

[0136] The above-described implementation method of this application involves inputting the operating condition of the traction motor to be tested into the second traction motor model to obtain the parameter signal of the traction motor to be tested. Then, the parameter signal of the traction motor to be tested is matched with the reference feature signal to obtain the matching result. Based on the matching result, the fault condition of the traction motor to be tested is determined. Automatic diagnosis of traction motor faults is achieved through simulation and signal comparison.

[0137] To ensure the accuracy of fault diagnosis, in some embodiments, the above S202 includes the following steps: S2021 to S2024.

[0138] S2021: Based on the differences between the stable operating state and the faulty operating state in the above simulation results, the corresponding signal residuals are obtained.

[0139] Signal residual refers to the difference between the reference signal under stable operating conditions and the actual signal under faulty operating conditions. This residual reflects the behavioral changes of the motor under different operating conditions and is a key characteristic of faults.

[0140] In the implementation, simulations are performed to acquire signal data of the motor under stable and faulty operating conditions. These signals may include current waveforms, vibration signals, temperature data, and speed curves. Under normal operating conditions, the motor exhibits a stable behavior pattern, while under fault conditions, the motor's behavior changes depending on the type of fault. Difference calculations are used to calculate the difference between each signal under faulty and normal conditions, obtaining the corresponding signal residuals.

[0141] S2022: Extract the time-frequency domain changes from the feature signals in the above simulation results to obtain the time-frequency domain change signal.

[0142] Time-frequency domain analysis is a technique that processes signals simultaneously in the time and frequency domains, revealing the spectral characteristics of a signal as it changes over time. In motor fault detection, changes in the time and frequency domains can provide more useful information, especially for non-stationary signals, where changes can more accurately reflect the dynamic characteristics at the time of fault occurrence.

[0143] In practical implementation, by performing time-frequency analysis methods such as short-time Fourier transform, wavelet transform, or Hilbert-Huang transform on the simulated motor signal, the changes of the signal in both time and frequency dimensions can be obtained, thus obtaining the time-frequency domain variation signal.

[0144] S2023: Extract the spatial variation from the feature signals in the above simulation results to obtain the spatial variation signal.

[0145] Spatial variation analysis focuses on the differences between signals collected from multiple sensor locations. For example, multiple sensors for a motor may be distributed at different locations within the motor. When a motor malfunctions, the signals collected by different sensors may exhibit different patterns of change. By analyzing these spatially distributed signals, spatial variation analysis helps determine the specific location or type of fault.

[0146] In the specific implementation, spatial variation features are extracted from the data collected by multiple sensors. Specifically, spatial variation can be extracted from the feature signals in the above simulation results through methods such as spatial filtering and spatial correlation analysis to obtain spatial variation signals.

[0147] S2024: Based on the above signal residual, the above time-frequency domain variation signal and the above spatial variation signal, obtain the reference characteristic signal corresponding to the preset fault condition.

[0148] In the specific implementation, the three types of signal features extracted in steps S2021, S2022, and S2023 are integrated: signal residual, time-frequency domain variation signal, and spatial variation signal. By combining these different types of feature signals, reference feature signals under the fault conditions are constructed. Each fault type has a specific set of reference feature signals, which are obtained based on multiple simulations and historical data. Through a multi-dimensional data fusion method, the corresponding reference feature signals under the preset fault conditions are obtained.

[0149] The above-described implementation method of this application obtains the corresponding signal residual based on the difference between the stable operating state and the fault operating state in the simulation results. Then, it extracts the time-frequency domain changes in the feature signals in the simulation results to obtain the time-frequency domain change signal. Furthermore, it extracts the spatial changes in the feature signals in the simulation results to obtain the spatial change signal. Finally, based on the signal residual, the time-frequency domain change signal, and the spatial change signal, it obtains the reference feature signal corresponding to the preset fault condition. By integrating these feature signals, a fault reference model is established, ensuring the accuracy and reliability of fault diagnosis.

[0150] In one embodiment of this application, the flowchart of the traction motor fault diagnosis system based on digital twin is as follows: Figure 3 As shown, the specific steps include: S1 to S10.

[0151] Step S1: Obtain the electrical, structural, and material parameters of the motor. For example, referencing the model of a three-phase squirrel-cage asynchronous traction motor for trains, obtain electrical parameters such as rated power, rated voltage, rated speed, rated frequency, rated current, rated torque, and number of pole pairs. Obtain geometric dimensions such as the number of stator slots, stator inner and outer diameters, stator core length, number of winding layers, number of wires per slot, coil pitch, number of rotor slots, air gap length, rotor inner diameter, and core length. Obtain the materials of components such as rotor rods, rotor rings, stator core, rotor core, housing, and shaft. Add eddy current effects to all rotor metal rods to induce current; add core loss and eddy current loss effects to the given rotor core for subsequent loss analysis; prepare for the construction and association of the electromagnetic thermal model of the traction motor.

[0152] Step S2: Establish the asynchronous traction motor model. Select a three-phase asynchronous squirrel-cage motor as the design type. Assign values ​​to the electrical, geometric, and material parameters of the traction motor to generate a finite element analysis model, and add eddy currents and losses. Refine the heat transfer parts of the motor model, such as the insulation layer, air gap, and air-cooled area of ​​the outer shell. Select and set the heat transfer parameters such as thermal conductivity, specific heat capacity, and density of the thermal model component materials. Divide the surface and volume meshes of the refined parts of the thermal model. Set the heat source and internal and external heat dissipation conditions accordingly. The external heat dissipation method of the cover is natural convection during test bench testing and forced convection during high-speed train operation. Adjust the heat transfer coefficient by setting it. The internal heat dissipation of the motor is forced convection, with air entering the ventilation port from the winding end and being blown out after passing through the ventilation channels inside the motor stator and rotor. An extended model can be the dynamic model of the traction motor.

[0153] Step S3: Operating condition limits for the traction motor. These limits mainly include: temperature limits (continuous operation, short-term overload), current limits (rated current, overload current), voltage limits (rated voltage, fluctuation range), speed limits (rated speed, maximum speed), vibration limits, and noise limits. These operating condition limits serve the data requirements for model order reduction. Traction motor model order reduction employs Equivalent Circuit Extraction (ECE) and Reduced Order Model (ROM) techniques. ECE technology scans motor current, rotor position, angle, and other information based on the motor type, calculating the torque, flux linkage, inductance, and other results corresponding to each scan point. ROM technology uses data-driven analysis to identify and extract the main characteristics of the motor's temperature distribution by analyzing a large amount of thermal model simulation data based on different operating conditions, parameter settings, or input conditions.

[0154] Step S4: Traction motor model reduction. The electromagnetic model of the traction motor is reduced using ECE technology. An external circuit for the motor's operating conditions is constructed, and the operating conditions within the limits are scanned at a certain density. Parameters such as current and voltage input to the motor model are set, as are output parameters such as motor torque, speed, and losses. Basic parameters such as motor resistance and variable parameter tables such as current and inductance changes are extracted. The thermal model of the traction motor is reduced using ROM technology. Through Design of Experiment (DOE), an experimental matrix is ​​created based on the range of input parameters (losses output by the motor's electromagnetic model) to obtain simulated parameter combinations and train the operating temperature field distribution. A certain proportion of the training data is selected to generate the reduced-order model, while unselected data is used to verify the accuracy of the reduced-order model. For the losses between the electromagnetic and thermal models, iron losses and synchronous motor magnetic losses are obtained through simulation using the reduced-order model. Copper losses are calculated based on the number of phases, the effective value of the current, and the resistance value of a single phase.

[0155] Step S5: Input of actual and historical motor data. This can be done by collecting actual operating data from the connected traction motor or by directly importing historical traction motor data recorded by sensors. The actual motor operating signals, such as current, contain a large amount of noise, requiring data noise reduction preprocessing, such as using moving average, wavelet transform, or other methods.

[0156] Step S6 involves correcting the traction motor model. Model correction is performed on the reduced-order motor model based on measured data. The corrected simulation model exhibits high fidelity and supports digital twin applications. Commonly used motor parameter identification algorithms include dynamic forgetting factor recursive least squares, extended Kalman filtering, and model reference adaptive methods. The parameter identification algorithm module identifies parameters from the input measured motor data. The identified stator resistance, inductance, and rotor flux linkage values ​​are injected as correction values ​​into the ECE model, ultimately ensuring that the ECE output approximates the measured results. Steps S5 and S6 correct the traction motor simulation model to form a digital twin model of the traction motor. The principle block diagram is shown below. Figure 4 As shown.

[0157] Step S7: Input fault conditions in the software. For stator faults with inter-turn short circuits, modify the corresponding percentage of turns in the traction motor digital twin electromagnetic model to simulate different degrees of inter-turn short circuit faults. For faults with loose core or air gap misalignment, modify the displacement of the corresponding core and air gap relative to their original positions in the traction motor digital twin model to simulate different degrees of core looseness and air gap misalignment. For bearing faults, in the traction motor digital twin electromagnetic model, coupling breakage is equivalent to a sudden decrease in motor load. For rotor bar breakage faults, the metal bars are made non-fixed in the traction motor digital twin model. For motor phase loss caused by winding short circuits, the corresponding open-circuit phase current is controlled in the traction motor digital twin electromagnetic model. In the expandable dynamic model, for bearing faults (wear, spalling, pitting, cracking, etc.), damping force is applied based on rolling frequency and orientation to simulate imperfect rolling caused by deformation.

[0158] Step S8, Traction Motor Digital Twin Model. In this step, a modified traction motor digital twin model is used to conduct simulation experiments under no-load, rated load, and software-set fault conditions.

[0159] Step S9, Traction Motor Characteristic Parameter Signals. Based on simulation experiments of no-load, rated load, and software-set fault conditions, characteristic signal data such as current, torque, speed, magnetic field, and temperature are read. This constructs multiple sets of characteristic signal outputs for stable operation of the motor system, along with signal residuals under different degrees of motor faults; it also obtains the time-frequency domain variations of these characteristic signals and the spatial distribution variations of the field characteristic signals.

[0160] Step S10: Traction motor fault diagnosis based on digital twins. By matching the residuals of different characteristic signals from the comprehensive model and considering the differences in time-frequency and spatial domain variations, fault classification and identification are performed. In the state space of characteristic signals, time-frequency domain, and spatial domain differences (residuals and variations), the difference responses of a fault at different degrees constitute the response spectrum of the fault mode. The traction motor fault diagnosis system based on digital twins provides the diagnostic method with a large amount of idealized characteristic signal data that is difficult for sensors to observe, which can greatly improve the accuracy of fault classification, identification, and diagnosis.

[0161] In the specific process of setting up a traction motor fault and obtaining the corresponding characteristic signals, when simulating a phase loss fault in the traction motor, based on the rated operating conditions of the traction motor, a resistor of a certain value is added to the corresponding phase circuit to simulate different degrees of current loss in that phase circuit. The abnormal signal caused by the fault is: a severe imbalance in the stator rotating magnetic field, resulting in a negative sequence current in the stator. The negative sequence magnetic field and the rotor induce an electromotive force, causing a surge in rotor current, severe rotor heating, and intensified motor vibration. Fault characteristic signal illustration: when the added resistance is large enough, the open-circuit phase current is 0, and the stable current amplitude of the other two phases increases from approximately 120A to approximately 400A, representing sinusoidal currents with a 120° phase difference. (Reference) Figure 5 and Figure 6 , Figure 5 The current curves of the three phases of the motor during a phase loss fault are shown. Figure 6 This is a comparison of the non-open-circuit current of a motor with a phase loss fault and a normal motor.

[0162] When simulating a traction motor rotor lock-up fault, based on the traction motor's rated operating conditions, a mass block with a certain moment of rotational inertia is added to the load to simulate different degrees of rotor lock-up. Abnormal signals caused by the fault: When the motor is stalled, it cannot generate normal back electromotive force, the current increases sharply, the temperature rises rapidly, and the motor's magnetic field becomes static, distinguishing it from a single-phase fault. Fault characteristic signal illustration: When the added mass block's moment of rotational inertia is sufficiently large, the motor speed drops to 0, the amplitude of the three-phase current increases from approximately 120A under rated conditions to approximately 500A, and the stator copper loss approximately doubles. (Reference) Figure 7 and Figure 8 , Figure 7 The curves showing the changes in the three-phase current of the motor during a stall fault; Figure 8 The curve shows the change in stator copper loss during a stall fault.

[0163] When simulating a coupling breakage fault, based on the rated operating conditions of the traction motor, the motor load is controlled to decrease stepwise at the corresponding breakage moment to simulate different degrees of coupling breakage. Abnormal signals caused by the fault: Coupling breakage is usually accompanied by collision or friction between metal parts, resulting in abnormal vibration and noise. Motor torque oscillation decreases, current decreases, copper losses decrease, heat dissipation decreases, and temperature drops. Fault characteristic signal illustration: When simulating complete coupling breakage, the motor load abruptly drops to 0, the current amplitude decreases from approximately 120A to approximately 30A, both maintained as sinusoidal currents with a 120° phase difference. Stator copper losses are significantly reduced, and the motor thermal field cools significantly. (Reference) Figure 9 and Figure 10 , Figure 9 The current curves of the three phases of the motor before and after the coupling completely breaks; Figure 10 A comparison of the changes in the motor's thermal field before and after the coupling completely breaks.

[0164] When simulating mechanical jamming faults, based on the traction motor's no-load condition, the motor load is increased to simulate different degrees of mechanical jamming. Abnormal signals caused by the fault include: increased motor torque, increased current, slight decrease in speed, and increased temperature.

[0165] When simulating inter-turn short-circuit faults, based on the rated operating conditions of the traction motor, the number of turns in the corresponding winding slots is set to simulate different degrees of inter-turn short-circuit faults. Abnormal signals caused by the fault include: increased short-circuit turn current, magnetic field asymmetry, air gap magnetic field distortion, and motor current imbalance due to changes in induced electromotive force. Fault characteristic signals are illustrated as follows: phase currents gradually increase with the severity of the short-circuit fault; non-short-circuit phase currents also slightly increase due to the influence of inter-phase magnetic flux coupling; the phase difference between non-short-circuit phase currents gradually decreases, while the phase difference between non-short-circuit phase currents and short-circuit phase currents gradually increases.

[0166] Compared to the method described above, which calculates the difference between the stable solution of the characteristic signal and the fault condition under the ideal assumptions of the mathematical mechanism model, the characteristic signal data calculated by the electromagnetic model simulation of the traction motor is a time series. Furthermore, after parameter identification and model correction, the simulated characteristic signal values ​​of the traction motor digital twin model are theoretically unbiased compared to the actual motor data. Typically, sensors can only provide point data such as current, voltage, torque, vibration, and temperature, and are dependent on a motor that can still operate normally after a fault. In contrast, the traction motor digital twin model can provide field data such as magnetic flux and temperature, more intuitively reflecting the different impacts of traction motor faults on different locations and components of the motor. The traction motor digital twin model has rich fault injection mechanisms, allowing for the expansion of fault types input into the software through modifications to the mechanical model, model parameters, simulation settings, and the actual model without damaging the hardware. Based on the multi-characteristic signal data and time-frequency and spatial domain variations of the traction motor digital twin model, the traction motor fault response spectrum is richer, and the accuracy of various fault diagnosis algorithms is generally improved.

[0167] Based on the motor model construction method provided in the above embodiments, this application also provides specific implementation methods of the motor model construction device. Please refer to the following embodiments.

[0168] First see Figure 11 The motor model building device 110 provided in this application embodiment includes the following modules:

[0169] The acquisition module 111 is used to acquire the physical parameters of the target traction motor.

[0170] Module 112 is used to construct the corresponding finite element analysis model based on the above-mentioned motor physical parameters.

[0171] The meshing module 113 is used to mesh the finite element analysis model according to the heat transfer parameters of each component in the finite element analysis model to obtain the heat transfer model.

[0172] The simulation module 114 is used to perform heat transfer simulation according to the above heat transfer model to obtain the corresponding thermal characteristics.

[0173] The determination module 115 is used to obtain the first traction motor model based on the above finite element analysis model and the corresponding thermal characteristics.

[0174] The model reduction module 116 is used to reduce the model order of the first traction motor model according to the preset operating condition threshold of the target traction motor to obtain the second traction motor model.

[0175] The motor model construction device provided in this application obtains the physical parameters of the target traction motor, constructs a corresponding finite element analysis model based on the physical parameters, meshes the finite element analysis model according to the heat transfer parameters of each part in the finite element analysis model to obtain a heat transfer model, and then performs heat transfer simulation according to the heat transfer model to obtain the corresponding thermal characteristics. Based on the finite element analysis model and the corresponding thermal characteristics, a first traction motor model is obtained. Then, according to the operating condition threshold, the order of the first traction motor model is reduced to obtain a second traction motor model. By simulating the behavior of the motor under various operating conditions and comprehensively considering the physical characteristics of the motor in different aspects, a high-precision digital twin model is constructed, thereby accurately and comprehensively describing the characteristics of the entire motor.

[0176] As one implementation of this application, the motor model building device 110 further includes:

[0177] The determination module is used to obtain the actual operating data of the target traction motor based on its operating status.

[0178] The identification module is used to identify parameters in the above-mentioned actual operating data according to a preset parameter identification algorithm, and to determine the key parameters in the above-mentioned actual operating data.

[0179] The determination module is also used to obtain the corresponding correction values ​​based on the aforementioned key parameters.

[0180] The correction module is used to correct the second traction motor model according to the correction values ​​mentioned above.

[0181] As one implementation of this application, the aforementioned determining module includes:

[0182] The simulation unit is used to simulate the key parameters based on the first traction motor model described above, and to obtain simulation results corresponding to the key parameters.

[0183] The comparison unit is used to compare the above simulation results with the above key parameters to obtain the simulation error.

[0184] The determining unit is used to obtain the corresponding correction value based on the preset parameter identification algorithm and the above simulation error.

[0185] As one implementation of this application, the modified module includes:

[0186] The replacement unit is used to replace the corresponding parameter values ​​in the second traction motor model according to the above-mentioned correction values.

[0187] The calibration unit is used to recalibrate the second traction motor model after the parameter values ​​of the second traction motor model have been replaced.

[0188] As one implementation of this application, the defined module includes:

[0189] The acquisition unit is used to collect data related to the operating status of the target traction motor.

[0190] The noise reduction unit is used to perform noise reduction preprocessing on the above-mentioned operating status-related data to obtain the actual operating data of the target traction motor.

[0191] As one implementation of this application, the order reduction module 116 includes:

[0192] The acquisition unit is used to acquire the preset electrical operating condition threshold and thermal operating condition threshold of the target traction motor.

[0193] The order reduction unit is used to reduce the order of the electromagnetic model of the first traction motor model by extracting it using an equivalent circuit based on the electrical characteristics of the first traction motor model.

[0194] The order reduction unit is also used to reduce the order of the heat transfer model of the first traction motor model based on the thermal characteristics of the first traction motor model, so as to obtain the second traction motor model.

[0195] As one implementation of this application, the motor model building device 110 further includes:

[0196] The simulation module is used to input preset fault conditions into the second traction motor model, and to conduct simulation experiments through the second traction motor model to obtain simulation results.

[0197] The extraction module is used to extract the reference feature signal corresponding to the preset fault condition from the above simulation results.

[0198] As one implementation of this application, the motor model building device 110 further includes:

[0199] The determination module is used to input the operating conditions of the traction motor to be tested into the second traction motor model mentioned above, and obtain the parameter signals of the traction motor to be tested.

[0200] The determination module is also used to match the parameter signals of the traction motor to be tested with the reference feature signals to obtain a matching result.

[0201] The determination module is also used to determine the fault status of the traction motor to be tested based on the above matching results.

[0202] As one implementation of this application, the extraction module includes:

[0203] The determination unit is used to obtain the corresponding signal residual based on the difference between the stable operating state and the faulty operating state in the above simulation results.

[0204] The determination unit is also used to extract the time-frequency domain changes in the feature signals from the above simulation results to obtain the time-frequency domain change signal.

[0205] The extraction unit is used to extract the spatial changes in the feature signals from the above simulation results to obtain the spatial change signal.

[0206] The determining unit is also used to obtain the reference characteristic signal corresponding to the preset fault condition based on the above-mentioned signal residual, the above-mentioned time-frequency domain change signal and the above-mentioned spatial change signal.

[0207] Each module in the vehicle trajectory planning device provided in this application embodiment can implement each step in the above-mentioned vehicle trajectory planning method and achieve the corresponding effect. For the sake of brevity, it will not be described in detail here.

[0208] Figure 12 A schematic diagram of the vehicle trajectory planning hardware provided in an embodiment of this application is shown.

[0209] The vehicle trajectory planning device may include a processor 1201 and a memory 1202 storing computer program instructions.

[0210] Specifically, the processor 1201 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0211] Memory 1202 may include mass storage for data or instructions. For example, and not limitingly, memory 1202 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1202 may include removable or non-removable (or fixed) media. Where appropriate, memory 1202 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 1202 is non-volatile solid-state memory.

[0212] The memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the vehicle trajectory planning method according to any embodiment of this disclosure.

[0213] The processor 1201 reads and executes computer program instructions stored in the memory 1202 to implement any of the vehicle trajectory planning methods in the above embodiments.

[0214] In one example, the vehicle trajectory planning device may also include a communication interface 1203 and a bus 1210. For example, Figure 12 As shown, the processor 1201, memory 1202, and communication interface 1203 are connected through bus 1210 and complete communication with each other.

[0215] The communication interface 1203 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0216] Bus 1210 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 1210 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0217] Furthermore, in conjunction with the methods for constructing motor models in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the motor model construction methods in the above embodiments.

[0218] This application also provides a computer program product, including a computer program that, when executed, implements any of the methods for constructing a motor model as described in the above embodiments.

[0219] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0220] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0221] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in 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, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams.

[0222] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for constructing a motor model, characterized in that, The method includes: Obtain the physical parameters of the target traction motor; Based on the physical parameters of the motor, a corresponding finite element analysis model is constructed; According to the heat transfer parameters of each component in the finite element analysis model, the finite element analysis model is meshed to obtain the heat transfer model; According to the heat transfer model, heat transfer simulation is performed to obtain the corresponding thermal characteristics; Based on the finite element analysis model and the corresponding thermal characteristics, the first traction motor model is obtained; Based on the preset operating condition threshold of the target traction motor, the first traction motor model is reduced in order to obtain the second traction motor model.

2. The motor model construction method according to claim 1, characterized in that, After reducing the order of the first traction motor model according to the preset operating condition threshold of the target traction motor to obtain the second traction motor model, the method further includes: Based on the operating status of the target traction motor, the actual operating data of the target traction motor is obtained; Based on a preset parameter identification algorithm, the actual operating data is subjected to parameter identification to determine the key parameters in the actual operating data. Based on the key parameters, the corresponding correction values ​​are obtained; The second traction motor model is modified according to the aforementioned correction value.

3. The motor model construction method according to claim 2, characterized in that, The step of obtaining the corresponding correction value based on the key parameters includes: Based on the first traction motor model, simulations are performed on the key parameters to obtain simulation results corresponding to the key parameters; The simulation results are compared with the key parameters to obtain the simulation error; Based on the preset parameter identification algorithm and the simulation error, the corresponding correction value is obtained.

4. The motor model construction method according to claim 2, characterized in that, The step of correcting the second traction motor model according to the correction value includes: Replace the corresponding parameter values ​​in the second traction motor model according to the correction values; Once the parameter values ​​of the second traction motor model have been replaced, the second traction motor model is recalibrated.

5. The method for constructing a motor model according to claim 2, characterized in that, The process of obtaining the actual operating data of the target traction motor based on its operating status includes: Collect relevant data on the operating status of the target traction motor; The noise reduction preprocessing of the operating status-related data is performed to obtain the actual operating data of the target traction motor.

6. The method for constructing a motor model according to claim 1, characterized in that, The target traction motor's physical parameters include at least one of electrical parameters, geometric parameters, and material parameters; The step of reducing the order of the first traction motor model to obtain the second traction motor model according to the preset operating condition threshold of the target traction motor includes: Obtain the preset electrical operating condition threshold and thermal operating condition threshold of the target traction motor; Based on the electrical characteristics of the first traction motor model, the electromagnetic model of the first traction motor model is reduced in order using equivalent circuit extraction. Based on the thermal characteristics of the first traction motor model, the heat transfer model of the first traction motor model is reduced in order using a reduced-order model to obtain the second traction motor model.

7. The method for constructing a motor model according to any one of claims 1 to 6, characterized in that, After reducing the order of the first traction motor model according to the preset operating condition threshold of the target traction motor to obtain the second traction motor model, the method further includes: The preset fault conditions are input into the second traction motor model, and simulation experiments are conducted through the second traction motor model to obtain simulation results; Extract the reference feature signal corresponding to the preset fault condition from the simulation results.

8. The method for constructing a motor model according to claim 7, characterized in that, After extracting the reference feature signal corresponding to the preset fault condition from the simulation results, the method further includes: The operating conditions of the traction motor to be tested are input into the second traction motor model to obtain the parameter signals of the traction motor to be tested; The parameter signal of the traction motor to be tested is matched with the reference feature signal to obtain the matching result; Based on the matching results, the fault condition of the traction motor to be tested is determined.

9. The method for constructing a motor model according to claim 7, characterized in that, Extracting the reference feature signal corresponding to the preset fault condition from the simulation results includes: Based on the differences between the stable operating state and the faulty operating state in the simulation results, the corresponding signal residuals are obtained; Extract the time-frequency domain changes from the feature signals in the simulation results to obtain the time-frequency domain change signal; The spatial variation signal is obtained by extracting the spatial variation from the feature signals in the simulation results; Based on the signal residual, the time-frequency domain variation signal, and the spatial variation signal, a reference characteristic signal corresponding to the preset fault condition is obtained.

10. A motor model building device, characterized in that, The device includes: The acquisition module is used to acquire the physical parameters of the target traction motor. The construction module is used to construct the corresponding finite element analysis model based on the physical parameters of the motor; The meshing module is used to mesh the finite element analysis model according to the heat transfer parameters of each component in the finite element analysis model, so as to obtain a heat transfer model. The simulation module is used to perform heat transfer simulation according to the heat transfer model to obtain the corresponding thermal characteristics; The determination module is used to obtain the first traction motor model based on the finite element analysis model and the corresponding thermal characteristics; The model reduction module is used to reduce the model order of the first traction motor model according to the preset operating condition threshold of the target traction motor, so as to obtain the second traction motor model.

Images