Motor optimization method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a motor optimization method and device, electronic equipment and a storage medium, and the method comprises the steps: building a response surface model of each physical field according to the working condition parameters corresponding to a plurality of working condition sampling points and the reference response values of various physical fields under each working condition sampling point; utilizing the response surface model of each physical field to predict an initial response value of an individual in the initial population in each physical field based on the initial population; based on the optimization function corresponding to the motor optimization target, the initial population and the initial response value of the individual in the initial population in each physical field, optimization calculation is carried out, an optimization result is obtained, and the optimization result comprises the optimization value of the optimization variable in the optimization function and the optimized response surface model of at least one physical field. By adopting the method, the optimization efficiency of the motor can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to a motor optimization method, device, electronic device, and storage medium. Background Art

[0002] In the related art, the motor design process is usually simulated using simulation software to select the solution with the best simulation effect. However, when using simulation software for optimization, due to the slow simulation speed of the simulation software, it may take several weeks. Therefore, the motor optimization design in the related art has the problem of low optimization efficiency. Summary of the Invention

[0003] In view of this, the embodiments of the present application propose a motor optimization method, device, electronic device and storage medium, which can effectively improve the efficiency of motor optimization design by combining response surface models of multiple physical fields to optimize the motor instead of the simulation scheme.

[0004] In a first aspect, an embodiment of the present application provides a motor optimization method, the method comprising: obtaining an initial population, a motor optimization target, and a sample operation data set, the sample operation data set comprising operating parameters corresponding to a plurality of operating sampling points and respective reference response values ​​of a plurality of physical fields of the motor at each operating sampling point; establishing a response surface model for each physical field based on the operating parameters corresponding to the plurality of operating sampling points and respective reference response values ​​of a plurality of physical fields at each operating sampling point; utilizing the response surface model of each physical field to predict the initial response values ​​of individuals in the initial population in each physical field based on the initial population; performing optimization calculations based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of individuals in the initial population in each physical field to obtain optimization results, the optimization results comprising the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

[0005] In the second aspect, an embodiment of the present application provides a motor optimization device, including a data acquisition module for acquiring an initial population, a motor optimization target, and a sample operation data set obtained using simulation software, wherein the sample operation data set includes operating parameters corresponding to multiple operating condition sampling points and reference response values ​​of multiple physical fields at each operating condition sampling point output based on the operating condition parameters of each operating condition when the simulation software performs simulation; a model establishment module for establishing a response surface model for each physical field based on the operating parameters corresponding to the multiple operating condition sampling points and the reference response values ​​of multiple physical fields at each operating condition sampling point; a response value prediction module for using the response surface model of each physical field to predict the initial response values ​​of individuals in the initial population in each physical field based on the initial population; an optimization calculation module for performing optimization calculation based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of individuals in the initial population in each physical field to obtain optimization results, wherein the optimization results include the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

[0006] In one embodiment, the sample operation data set includes a training sample set; the model building module includes a simulation prediction submodule, a loss calculation submodule and a parameter adjustment submodule, the simulation prediction submodule is used to obtain the operating condition parameters corresponding to at least one operating condition sampling point from the training sample set for the target physical field, and use the Kriging model to perform simulation analysis based on the operating condition parameters to obtain the simulation response value corresponding to the operating condition parameter under the target physical field, and the target physical field is any one of a plurality of physical fields; the loss calculation submodule is used to obtain the simulation analysis loss based on the simulation response value and the reference response value corresponding to the simulation response value; the parameter adjustment module is used to adjust the model parameters of the Kriging model based on the simulation analysis loss, and use the Kriging model after the last adjustment of the model parameters as the response surface model of the target physical field.

[0007] In one embodiment, the sample operation data set also includes a verification sample set; the motor optimization device also includes a verification module, which is used to obtain the operating condition parameters corresponding to at least one operating condition sampling point from the verification sample set for the target physical field, and use the response surface model corresponding to the target physical field to perform simulation analysis based on the operating condition parameters to obtain the verification response value corresponding to the operating condition parameter under the target physical field; and obtain the verification analysis loss based on the verification response value and the reference response value corresponding to the verification response value.

[0008] In one embodiment, the model building module includes a standardization processing submodule, a normalization processing submodule and a model building submodule; the standardization processing submodule is used to standardize the operating condition parameters to obtain the standardized operating condition parameters; the normalization processing submodule is used to normalize the standardized operating condition parameters to obtain the normalized operating condition parameters; the model building submodule is used to establish a response surface model for each physical field based on the reference response value of each physical field under the action of the multiple operating condition sampling points and the normalized operating condition parameters corresponding to the multiple operating condition sampling points.

[0009] In one possible implementation, the data acquisition module is further configured to acquire constraints and decision variables of the motor; and obtain an optimization function of the target motor according to the optimization target, the constraints and the optimization variables.

[0010] In one embodiment, the multiple physical fields include electromagnetic fields, thermal fields, and stress fields.

[0011] In one embodiment, the data acquisition module is further used to obtain operating parameters of multiple operating sampling points; and use simulation software to output reference response values ​​of multiple physical fields of the motor at each operating sampling point based on the operating parameters of each operating condition.

[0012] In one embodiment, the optimization calculation module is also used to perform optimization calculations based on the optimization function corresponding to the motor optimization objective, the initial population, the initial response values ​​of individuals in the initial population in each physical field, and the response surface model of at least one physical field using a multi-objective optimization algorithm to obtain optimized values ​​of the optimization variables in the optimization function and an optimized response surface model of at least one physical field.

[0013] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor and a memory; one or more programs are stored in the memory and configured to be executed by the processor to implement the above method.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which program code is stored, wherein the above method is executed when the program code is executed by a processor.

[0015] The embodiments of the present application provide a motor optimization method, device, electronic device and storage medium. By establishing a response surface model for each physical field based on the operating parameters corresponding to the multiple operating sampling points and the reference response values ​​of the multiple physical fields at each operating sampling point, the response surface model of each physical field is used to predict the initial response value of the individuals in the initial population in each physical field based on the initial population; and performing optimization calculations based on the optimization target, the initial population and the initial response value of the individuals in the initial population in each physical field to obtain the optimization result. The response surface model can be used to replace time-consuming simulations, shortening the traditional design cycle of several weeks to several days or even hours, thereby greatly improving the efficiency of motor optimization simulation. In addition, through multi-objective collaborative optimization, factors such as electromagnetic, thermal management and structural strength in the operation of the motor are fully considered, and multiple objectives can be balanced at the same time. The motor finally designed is more superior in performance and can better meet the needs of complex applications. The invalid search space and computing resource consumption are significantly reduced, thereby improving the efficiency of motor performance optimization design. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A schematic diagram of a flow chart of a motor optimization method provided in an embodiment of the present application is shown; Figure 2 Shown Figure 1 Flow chart of step S120; Figure 3 Another schematic diagram of a motor optimization method provided by an embodiment of the present application is shown; Figure 4 Another schematic diagram of a flow chart of a motor optimization method provided in an embodiment of the present application is shown; Figure 5 A schematic diagram of a response surface model provided in an embodiment of the present application is shown; Figure 6 Another schematic diagram of a motor optimization method provided by an embodiment of the present application is shown; Figure 7 A schematic diagram of the process flow of the optimization phase in a motor optimization method provided in an embodiment of the present application is shown; Figure 8 A connection block diagram of a motor optimization device proposed in an embodiment of the present application is shown; Figure 9A structural block diagram of an electronic device for executing the method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0018] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0019] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0020] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0021] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0022] It should be noted that the term "plurality" used in this document refers to two or more. "And / or" describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. The character " / " generally indicates an "or" relationship between the associated objects.

[0023] Figure 1 The motor optimization method of the present application is specifically shown. The method can be applied to an electronic device, which can be a terminal device or a server. The terminal device can be a mobile terminal or a vehicle, etc. The method includes: Step S110: Obtain an initial population, motor optimization targets, and a sample operation data set.

[0024] The sample operation data set includes operating parameters corresponding to a plurality of operating sampling points and respective reference response values ​​of a plurality of physical fields of the motor at each operating sampling point.

[0025] The initial population is the starting set of design parameters for the optimization algorithm. It typically includes multiple individuals, each representing a set of possible motor design solutions and containing specific values ​​for all optimization variables to be optimized. For example, this may include specific values ​​for one or more of the motor's permanent magnet width, number of winding turns, number of slots per pole and per phase, rotor pole arc coefficient, and air gap length.

[0026] The above-mentioned initial population can be obtained by sampling based on the existing value range of the optimization variable (such as Latin hypercube sampling or random sampling); it can also be extracted from sample operation data, which is not specifically limited here.

[0027] Motor optimization objectives refer to performance indicators that need to be maximized or minimized. These are typically multi-objective (conflicting) objectives, which can be of different types. These objectives may include electromagnetic performance (specific optimization objectives may include maximizing efficiency to improve electrode energy conversion efficiency and minimizing torque ripple to improve motor operating smoothness), thermal performance (specific optimization objectives may include minimizing the maximum operating temperature of the motor to prevent overheating failure), and structural performance (specific optimization objectives may include minimizing the maximum stress of the material to prevent material aging failure). The above optimization objectives are merely illustrative; more optimization objectives are possible and are not specifically limited.

[0028] Operating condition sampling points are representative operating state combinations selected during the motor multiphysics optimization process to cover the motor's actual operating range. Each sampling point corresponds to a specific set of input conditions (operating condition parameters). These parameters can include the motor's electrical parameters (e.g., load current, power supply voltage), mechanical parameters (e.g., motor speed, load torque), and environmental parameters (e.g., ambient temperature, humidity).

[0029] The reference response values ​​of the multiple physical fields at each operating condition sampling point may be obtained by operating the motor under the operating condition parameters at the operating condition sampling point to measure corresponding physical field data.

[0030] Specifically, the electrical parameters, mechanical parameters and environmental parameters of the motor can be adjusted to obtain the operating parameters corresponding to the operating sampling points, and the physical field data (such as electromagnetic field data (such as magnetic flux, current density, etc.), thermal field data (such as temperature distribution, heat flow, etc.) and rotor stress field data (such as stress distribution, strain, etc.) of the motor when it is running under the operating parameters corresponding to each operating sampling point are collected. Electromagnetic field data (such as magnetic flux, current density, etc.), thermal field data (such as temperature distribution, heat flow, etc.) and rotor stress field data (such as stress distribution, strain, etc.)

[0031] Another way to obtain the reference response values ​​of multiple physical fields at each operating sampling point is to use simulation software to generate corresponding physical field data such as electromagnetic field, thermal field, and stress field based on the operating parameters of each operating sampling point, that is, to generate electromagnetic field data, thermal field data, and rotor stress field data, etc.

[0032] In this case, the above-mentioned method of obtaining sample operation data may include: The operating parameters of multiple operating condition sampling points are obtained; and the simulation software is used to output the reference response values ​​of the motor's various physical fields at each operating condition sampling point based on the operating parameters of each operating condition.

[0033] Specifically, the operating parameters of multiple operating condition sampling points can be set based on experience, or representative operating condition sampling points can be selected in the operating condition parameter space (such as speed, load, ambient temperature, etc.) by using Latin hypercube sampling or full factorial design to obtain the operating condition parameters of the operating condition sampling point pair.

[0034] Specifically, the simulation software may be finite element analysis software, such as ANSYS Maxwell (ANSYS Maxwell Electromagnetic Field Simulation Software), JMAG (JMAG Electromagnetic Analysis Software), or COMSOL Multiphysics (COMSOL Multiphysics Modeling and Simulation Software), etc.

[0035] The above-mentioned use of simulation software to output the reference response values ​​of multiple physical fields of the motor at each working condition sampling point based on the working condition parameters of each working condition can be to establish a multi-physical field coupling model in the simulation software, so as to use the simulation software to input the working condition parameters into the multi-physical field coupling model to calculate the reference response values ​​of multiple physical fields at each working condition sampling point.

[0036] Step S120: establishing a response surface model for each physical field according to the operating parameters corresponding to the plurality of operating sampling points and the respective reference response values ​​of the plurality of physical fields at each operating sampling point.

[0037] Considering that different samples are collected in different measurement methods, if the response surface model is directly established using the collected samples, the resulting response surface model may be inaccurate. In addition, different operating parameters have large differences in parameters, and the amount of calculation is large when establishing the response surface model. Based on this, in one embodiment of the present application, the above step S120 includes: The operating condition parameters are standardized to obtain standardized operating condition parameters. The standardized operating condition parameters are normalized to obtain normalized operating condition parameters. A response surface model for each physical field is established based on the reference response value of each physical field under the action of the multiple operating condition sampling points and the normalized operating condition parameters corresponding to the multiple operating condition sampling points.

[0038] By adopting the above method steps, the unit symbols of all operating parameters can be unified in format. Subsequently, the operating parameters can be converted into a distribution with a mean of zero and a standard deviation of 1, thereby eliminating the dimensional differences of different parameters and avoiding inaccurate model training caused by parameters with too large a numerical range dominating the model training.

[0039] Among them, by standardizing and normalizing the operating parameters of multiple operating sampling points, the dimensional effect can be eliminated, the nonlinear fitting can be enhanced, and the model convergence can be accelerated, so as to ultimately improve the efficiency and accuracy of the subsequent response surface model of the physical field.

[0040] When establishing a response surface model, Kriging model, neural network model, regression model, etc. can be used as core modeling tools.

[0041] In one possible implementation, if the Kriging model is used as the core modeling tool, the above-mentioned step S120 can specifically be: using the operating parameters corresponding to each of the operating condition sampling points and the reference response values ​​of multiple physical fields at each operating condition sampling point, the Kriging model is trained to obtain a response surface model for each physical field.

[0042] The sample operation data set may specifically include a training sample set. The step of training the Kriging model using the operating parameters corresponding to each operating sampling point and the reference response values ​​of the multiple physical fields at each operating sampling point to obtain a response surface model for each physical field may include: Step S122: For the target physical field, obtain the operating condition parameters corresponding to at least one operating condition sampling point from the training sample set, use the Kriging model to perform simulation analysis based on the operating condition parameters, and obtain the simulation response value corresponding to the operating condition parameter under the target physical field, where the target physical field is any one of a plurality of physical fields.

[0043] Among them, if the target physical field is an electromagnetic field, the output simulation response value is magnetic density or efficiency; if the target physical field is a temperature field, the output simulation response value is temperature; if the target physical field is a stress field, the output simulation response value is stress value.

[0044] Step S124: obtaining a simulation analysis loss based on the simulation response value and a reference response value corresponding to the simulation response value.

[0045] Specifically, the simulation response value and the reference response value corresponding to the simulation response value may be calculated using an error calculation method such as mean square error or relative error, and the simulation analysis loss may be determined based on at least one of the calculated mean square error, relative error, etc.

[0046] Step S126: Adjust the model parameters of the Kriging model based on the simulation analysis loss.

[0047] If the training end condition is not met, return to the step of executing the operating condition parameters corresponding to at least one operating condition sampling point in the training sample set until the training end condition is met, and use the Kriging model after the last adjustment of the model parameters as the response surface model of the target physical field.

[0048] Specifically, the model parameters of the Kriging model are adjusted based on the simulation analysis loss to minimize the simulation analysis loss until the simulation analysis loss is less than a preset loss threshold or the number of iterative training reaches a preset number of iterations. It is determined that the iteration end condition is met, and the Kriging model after the last adjustment of the model parameters is used as the response surface model of the target physical field.

[0049] For example, in one round of iterative training, given n working condition sampling points and its response , for the input variables , the response value of the Kriging model It can be expressed as: Where, is a q-dimensional polynomial basis function, which is in the form of , is a positive integer; is the regression model parameter vector; The random error term that models the local deviation has a mean of zero and a variance of , the covariance matrix is: In the formula is the correlation function matrix. When adjusting model parameters, you can adjust the weights of the input variables. For example, if a certain operating parameter (such as speed) has a significant impact on temperature prediction, increase the weight corresponding to speed. You can also adjust related parameters in the covariance matrix.

[0050] It is worth mentioning that when the core modeling tool is a neural network model or a regression model, its model training process is similar to the aforementioned steps S122-S126, and will not be repeated here.

[0051] Step S130: using the response surface model of each physical field based on the initial population, predicting the initial response values ​​of individuals in the initial population in each physical field.

[0052] Specifically, when using established response surface models such as electromagnetic fields, thermal fields, and stress fields to predict the performance of each individual in the initial population (i.e., a set of operating parameters), the corresponding response surface model is called for each physical field, and the individuals in the initial population are input into the response surface model to output the initial response values ​​corresponding to the individuals in the initial population.

[0053] Step S140: Perform optimization calculations based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of the individuals in the initial population in each physical field to obtain optimization results, which include optimized values ​​of the optimization variables in the optimization function and an optimized response surface model of at least one physical field.

[0054] The purpose of the optimization calculation is to achieve an optimal trade-off among multiple optimization objectives (e.g., at least two of efficiency improvement, temperature reduction, and cost reduction) based on the multi-physical field response values ​​of the initial population. For example, a vector function can be constructed based on the multiple optimization objectives, and constraints can be set for each optimization objective. Subsequently, an iterative search is performed based on the initial population and the initial response values ​​of the individuals in the initial population in each physical field to find an optimization result that includes an optimal solution set for the multiple optimization objectives, and the optimization result satisfies the constraints.

[0055] It is worth mentioning that the optimization function is used to indicate the relationship between the optimization variables and the optimization target. The above-mentioned optimization variables may include the above-mentioned working condition types. For example, the optimization variables x may include the number of motor pole pairs p, the number of slots per pole per phase q, the cross-sectional area of ​​the shaft A, the electric density J, the motor crack ratio , motor rotor pole arc coefficient , stator tooth flux density , magnetic flux density in the slot , air gap length δ, winding width d, number of winding turns N, permanent magnet width , permanent magnet thickness The optimization goal may be one or more of maximizing motor efficiency, minimizing magnetic flux density, minimizing cost, etc.

[0056] In one embodiment, the above-mentioned step S140 can be to use a multi-objective optimization algorithm (such as the multi-objective genetic algorithm of NSGA-III) to perform optimization calculations based on the optimization function corresponding to the motor optimization objective, the initial population, the initial response values ​​of the individuals in the initial population in each physical field, and the response surface model of at least one physical field to obtain the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

[0057] Specifically, the initial population can be stratified according to the Pareto dominance relationship. Subsequently, reference points are generated according to the optimization objective. Each individual in the initial population is associated with the nearest reference point based on the vertical distance between each individual and the reference point, and the number of individuals associated with each reference point is counted. Based on the stratification and the selected reference points, the initial population (and the response surface model of at least one physical field) is iteratively updated. During the iterative update process, individuals in the initial population can be genetically modified and mutated, and individuals with higher fitness can be selected to form a new population. When the maximum number of iterations is reached, the Pareto front solution set is output. This solution set is the optimization result, which includes the optimized values ​​of the optimization variables and the optimized response surface model of at least one physical field. The reference points are virtual anchor points uniformly distributed in the target space and are used to guide the distribution of the solution set.

[0058] By adopting the above-mentioned method of the present application, a response surface model of each physical field is established according to the operating parameters corresponding to the multiple operating sampling points and the reference response values ​​of the multiple physical fields at each operating sampling point; the response surface model of each physical field is used to predict the initial response value of the individuals in the initial population in each physical field based on the initial population; and the optimization calculation is performed based on the optimization target, the initial population and the initial response value of the individuals in the initial population in each physical field to obtain the optimization result. It is possible to replace time-consuming simulations with response surface models and shorten the traditional design cycle of several weeks to several days or even hours. Thereafter, through multi-objective collaborative optimization, factors such as electromagnetic, thermal management and structural strength in the operation of the motor are fully considered, and it is possible to achieve a simultaneous balance of multiple objectives. The motor finally designed is superior in performance and can better meet the needs of complex applications. It significantly reduces the invalid search space and computing resource consumption, thereby improving the efficiency of motor performance optimization design.

[0059] In one embodiment, the sample operation data set also includes a validation sample set. Figure 3 Before executing step S130, the method further includes: Step S150: For the target physical field, obtain the operating condition parameters corresponding to at least one operating condition sampling point from the verification sample set, use the response surface model corresponding to the target physical field to perform simulation analysis based on the operating condition parameters, and obtain the verification response value corresponding to the operating condition parameters under the target physical field.

[0060] Step S160: Obtaining a verification analysis loss based on the verification response value and a reference response value corresponding to the verification response value.

[0061] The specific implementation process of the above-mentioned step S150 and step S160 can refer to the specific description of steps S122-S124 in the above-mentioned embodiment, and will not be repeated here.

[0062] Step S170: Determine whether the verification analysis loss is less than a preset loss threshold.

[0063] If the verification analysis loss is less than the preset loss threshold, step S140 is executed; if the verification analysis value is not less than the preset loss threshold, the process returns to step S122.

[0064] By adopting the above method, the accuracy of each response surface model obtained can be effectively ensured, so that the optimization result obtained by the final optimization calculation is more accurate.

[0065] For example, when performing motor optimization, it includes a model building phase and an optimization phase: 1. Model building stage like Figure 4As shown in the figure, the specific model construction process is: first, determine the input parameters and output parameters of the Kriging model, and secondly, determine the construction parameters of the Kriging model; thereafter, collect simulation data as sample data. Specifically, finite element analysis software can be used for detailed simulation, and multiple working conditions can be designed, including the operating status of the motor under different loads, different speeds and different working environment temperatures. The system collects electromagnetic field data (such as magnetic flux, current density, etc.), thermal field data (such as temperature distribution, heat flow, etc.) and rotor stress field data (such as stress distribution, strain, etc.) during the simulation process, and standardizes and normalizes the obtained simulation data to ensure that it meets the needs of processing and analysis, and excludes outliers and noise data to improve the reliability of subsequent modeling results. The specific training process is that, for each physical field, the training samples in the sample data are used to train the Kriging model to obtain the response surface model for each physical field, and the test samples in the sample data are used to test the response surface model for each physical field to determine whether the test error is less than the preset threshold. If the test error is small, that is, the test is passed, the response surface model can be used to predict the initial response value of the individuals in the initial population in each physical field based on the initial population; thereby calculating the individual fitness, and then performing cross-mutation operations on the individuals to select individuals with high fitness to form a new population, so as to finally obtain the optimization result. If the test error of the response surface model is large, the training sample needs to be used again to train the response surface model. Finally, the result is as follows Figure 5 The response surface model shown is worth mentioning that the response surface model is only schematic. Different dimensions of different response surface models have different meanings and will not be described in detail here.

[0066] For example, Figure 6 As shown, taking various physical fields including electromagnetic fields, thermal fields and stress fields as examples, after collecting simulation data, the simulation data can be preprocessed and the preprocessed data can be used as sample data. Thereafter, the electromagnetic field training data in the sample data is used to train the Kriging model, thereby constructing an electromagnetic field response surface model, and the electromagnetic field verification data in the sample data is used to verify whether the electromagnetic field response surface model meets expectations. If not, it is necessary to re-acquire the electromagnetic field training data in the sample data to train the electromagnetic field response surface model. If it meets expectations, the thermal field training data in the sample data is used to train the Kriging model, thereby constructing a thermal field response surface model; and verify whether the thermal field response surface meets expectations. If not, it is necessary to re-acquire the thermal field training data in the sample data to train the thermal field response surface model; if it meets expectations, the stress field training data in the sample data is used to train the Kriging model, thereby constructing a stress field response surface model; and verify whether the stress field response surface meets expectations. If not, it is necessary to re-acquire the stress field training data in the sample data to train the stress field response surface model; if it meets expectations, enter the subsequent optimization stage for optimization.

[0067] 2. Optimization stage.

[0068] See also Figure 7 Taking the optimization goal of minimizing magnetic flux density and maximizing motor efficiency as an example, the optimization variables x related to the optimization goal may include the number of motor pole pairs p, the number of slots per pole per phase q, the cross-street area A of the rotating shaft, the current density J, the motor split ratio , motor rotor pole arc coefficient , stator tooth flux density , magnetic flux density in the slot , air gap length δ, winding width d, number of winding turns N, permanent magnet width , permanent magnet thickness One or more of the above.

[0069] When the optimization goal is to minimize the magnetic flux density M and maximize the motor efficiency η, the relationship between the magnetic flux density and each optimization variable and the relationship between the motor efficiency and each optimization variable are established. For example, the optimization variables are , at this time, the minimized magnetic density can be expressed as , the maximum motor efficiency can be expressed as .

[0070] Afterwards, constraints can be set. For example, the constraints can be expressed as: ; Wherein, J is the aforementioned electric secret, is the slot fill rate, is the initial slot filling rate, η is the motor efficiency, is the initial efficiency of the motor, φ is the current angle, is the initial current angle, is the maximum output power, is the maximum output power initially set, The above constraints are only for illustration purposes, and there are more ways to set constraints, which are not specifically limited in this embodiment.

[0071] Finally, the NSGA III genetic algorithm is used to iteratively optimize the initial response values ​​of the individuals in the initial population in each physical field, as well as the response surface model of at least one physical field, based on the optimization objective and constraints. A determination is made as to whether an iterative termination condition has been met. If the iterative termination condition has not been met, the aforementioned step S140 of generating a reference point based on the optimization objective is further performed to iteratively update (inheritance and mutation) the initial value population. If the iterative termination condition has been met, the algorithm is terminated, and an optimization result is obtained. The solution set of the optimization result is displayed on the Pareto front. The optimization result includes the optimized values ​​of the optimization variables in the optimized optimization function and the optimized response surface model of at least one physical field.

[0072] By adopting the above method, the response surface model is used instead of the finite element simulation software to solve the motor performance under different parameters, and the multi-objective optimization design of the motor is carried out in combination with the NSGA III genetic algorithm. Among them, the NSGA III genetic algorithm iteratively searches for the optimal design through operations such as selection, crossover and mutation. At the end of the optimization, the non-dominated solutions on each solution set will be extracted and displayed on the Pareto front to help designers choose the best solution. This method not only improves the efficiency and accuracy of motor design by integrating simulation technology and advanced algorithms, but also increases the ability to handle multi-objective problems. In addition, the method also enhances the visual understanding of various performance trade-offs during the design process. The invention is suitable for gradually improving the comprehensive performance of various types of motor products and has broad application prospects in the motor manufacturing industry.

[0073] It is worth mentioning that the above method can be widely used in industries such as mechanical engineering, electrical engineering and control system design. It is especially suitable for occasions requiring high optimization and complex design constraints, such as electric vehicles, industrial automation, high-precision machine tools, aerospace and renewable energy equipment. Motor design and manufacturing.

[0074] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0075] See also Figure 8Another embodiment of the present application provides a motor optimization device 200, including a data acquisition module 210, which is used to obtain an initial population, a motor optimization target, and a sample operation data set obtained by using simulation software, wherein the sample operation data set includes operating parameters corresponding to multiple operating condition sampling points and reference response values ​​of multiple physical fields at each operating condition sampling point output based on the operating parameters of each operating condition when the simulation software performs simulation; a model establishment module 220, which is used to establish a response surface model for each physical field based on the operating parameters corresponding to the multiple operating condition sampling points and the reference response values ​​of multiple physical fields at each operating condition sampling point; a response value prediction module 230, which is used to predict the initial response value of individuals in the initial population in each physical field based on the initial population using the response surface model of each physical field; an optimization calculation module 240, which is used to perform optimization calculation based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response value of individuals in the initial population in each physical field to obtain an optimization result, wherein the optimization result includes the optimized value of the optimization variable in the optimization function and the optimized response surface model of at least one physical field.

[0076] In one embodiment, the sample operation data set includes a training sample set; the model building module 220 includes a simulation prediction submodule, a loss calculation submodule and a parameter adjustment submodule, the simulation prediction submodule is used to obtain the operating condition parameters corresponding to at least one operating condition sampling point from the training sample set for the target physical field, and use the Kriging model to perform simulation analysis based on the operating condition parameters to obtain the simulation response value corresponding to the operating condition parameter under the target physical field, and the target physical field is any one of a plurality of physical fields; the loss calculation submodule is used to obtain the simulation analysis loss based on the simulation response value and the reference response value corresponding to the simulation response value; the parameter adjustment module is used to adjust the model parameters of the Kriging model based on the simulation analysis loss, and use the Kriging model after the last adjustment of the model parameters as the response surface model of the target physical field.

[0077] In one embodiment, the sample operation data set also includes a verification sample set; the motor optimization device 200 also includes a verification module, which is used to obtain the operating condition parameters corresponding to at least one operating condition sampling point from the verification sample set for the target physical field, and use the response surface model corresponding to the target physical field to perform simulation analysis based on the operating condition parameters to obtain the verification response value corresponding to the operating condition parameter under the target physical field; and obtain the verification analysis loss based on the verification response value and the reference response value corresponding to the verification response value.

[0078] In one embodiment, the model building module 220 includes a standardization processing submodule, a normalization processing submodule and a model building submodule; the standardization processing submodule is used to standardize the operating condition parameters to obtain the standardized operating condition parameters; the normalization processing submodule is used to normalize the standardized operating condition parameters to obtain the normalized operating condition parameters; the model building submodule is used to establish a response surface model for each physical field based on the reference response value of each physical field under the action of the multiple operating condition sampling points and the normalized operating condition parameters corresponding to the multiple operating condition sampling points.

[0079] In one embodiment, the data acquisition module 210 is further configured to acquire the constraints and decision variables of the motor; and obtain the optimization function of the target motor according to the optimization target, the constraints and the optimization variables.

[0080] In one embodiment, the multiple physical fields include electromagnetic fields, thermal fields, and stress fields.

[0081] In one embodiment, the data acquisition module 210 is further used to obtain operating parameters of multiple operating sampling points; and use simulation software to output reference response values ​​of multiple physical fields of the motor at each operating sampling point based on the operating parameters of each operating condition.

[0082] In one embodiment, the optimization calculation module 240 is also used to use a multi-objective optimization algorithm to perform optimization calculations based on the optimization function corresponding to the motor optimization objective, the initial population, the initial response values ​​of the individuals in the initial population in each physical field, and the response surface model of at least one physical field, to obtain the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

[0083] Each module in the above-mentioned device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules. It should be noted that the device embodiment in this application corresponds to the aforementioned method embodiment. The specific principles in the device embodiment can be found in the contents of the aforementioned method embodiment, which will not be repeated here.

[0084] The following will be combined Figure 9 An electronic device provided by this application is described.

[0085] See also Figure 9 Based on the motor optimization method provided in the above embodiment, the embodiment of the present application also provides another electronic device 100 including a processor 102 that can execute the above method. The electronic device 100 can be a vehicle.

[0086] The electronic device 100 further includes a memory 104 . The memory 104 stores a program capable of executing the contents of the aforementioned embodiments, and the processor 102 can execute the program stored in the memory 104 .

[0087] The processor 102 may include one or more cores for processing data and a message matrix unit. The processor 102 utilizes various interfaces and circuits to connect various components within the electronic device 100. It executes instructions, programs, code sets, or instruction sets stored in the memory 104, and accesses data stored in the memory 104 to perform various functions and process data within the electronic device 100. Optionally, the processor 102 may be implemented in hardware using at least one of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 102 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 102 and may be implemented as a separate communication chip.

[0088] In this embodiment, the processor 102 includes a main controller and a system-on-chip to implement the aforementioned method steps.

[0089] The memory 104 may include random access memory (RAM) or read-only memory (ROM). The memory 104 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 104 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, and instructions for implementing the various method embodiments described below. The data storage area may also store data acquired by the electronic device 100 during use.

[0090] The electronic device 100 may also include a network module and a screen. The network module is used to receive and transmit electromagnetic waves, converting them into electrical signals, thereby communicating with a communication network or other devices, such as a playback device. The network module may include various existing circuit components for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, memory, and the like. The network module can communicate with various networks such as the Internet, an intranet, or a wireless network, or with other devices via a wireless network. Such wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The screen can display interface content and perform data interaction, such as displaying the aforementioned interface and triggering operations via the screen.

[0091] The present application also provides a computer-readable storage medium that stores program code, which can be called by a processor to execute the method described in the above method embodiment.

[0092] The computer-readable storage medium may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Alternatively, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program code for executing any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code may be compressed, for example, in a suitable format.

[0093] The present application also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods described in the various optional implementations described above.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A motor optimization method, characterized in that: The method comprises: Obtaining an initial population, a motor optimization target, and a sample operation data set, wherein the sample operation data set includes operating parameters corresponding to a plurality of operating condition sampling points and reference response values ​​of a plurality of physical fields of the motor at each operating condition sampling point; Establishing a response surface model for each physical field according to the operating parameters corresponding to the plurality of operating sampling points and the reference response values ​​of the plurality of physical fields at each operating sampling point; The response surface model of each physical field is used to predict the initial response value of individuals in the initial population under each physical field based on the initial population; An optimization calculation is performed based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of the individuals in the initial population in each physical field to obtain an optimization result, which includes the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

2. The method according to claim 1, characterized in that The step of establishing a response surface model for each physical field based on the operating parameters corresponding to the plurality of operating sampling points and the reference response values ​​of the plurality of physical fields at each operating sampling point comprises: The Kriging model is trained using the operating parameters corresponding to each of the operating sampling points and the reference response values ​​of multiple physical fields at each operating sampling point to obtain a response surface model for each physical field.

3. The method according to claim 2, characterized in that The sample operation data set includes a training sample set; The method uses the operating parameters corresponding to each operating sampling point and the reference response values ​​of multiple physical fields at each operating sampling point to train the Kriging model to obtain a response surface model for each physical field, including: For a target physical field, obtaining operating condition parameters corresponding to at least one operating condition sampling point from the training sample set, performing simulation analysis based on the operating condition parameters using a Kriging model, and obtaining a simulation response value corresponding to the operating condition parameters under the target physical field, wherein the target physical field is any one of multiple physical fields; Obtaining a simulation analysis loss based on the simulation response value and a reference response value corresponding to the simulation response value; Adjust the model parameters of the Kriging model based on the simulation analysis loss, and return to the step of executing the operating parameters corresponding to at least one operating condition sampling point in the training sample set until the training end condition is reached, and use the Kriging model after the last adjustment of the model parameters as the response surface model of the target physical field.

4. The method according to claim 3, characterized in that The sample operation data set also includes a verification sample set; Before predicting the initial response value of the target motor corresponding to each physical field based on the initial population using the response surface model of each physical field, the method further includes: For the target physical field, obtain an operating condition parameter corresponding to at least one operating condition sampling point from the verification sample set, perform simulation analysis based on the operating condition parameter using a response surface model corresponding to the target physical field, and obtain a verification response value corresponding to the operating condition parameter under the target physical field; Obtaining a verification analysis loss based on the verification response value and a reference response value corresponding to the verification response value; If the verification analysis loss is less than a preset loss threshold, performing the step of using a response surface model of each physical field based on the initial population to predict an initial response value of the target motor corresponding to each physical field; If the verification analysis value is not less than the preset loss threshold, the process returns to the step of obtaining the operating condition parameters corresponding to at least one operating condition sampling point from the training sample set.

5. The method according to claim 1, wherein The step of establishing a response surface model for each physical field according to the reference response value of each physical field under the action of the multiple operating condition sampling points and the operating condition parameters corresponding to the multiple operating condition sampling points comprises: performing standardization processing on the operating condition parameters to obtain standardized operating condition parameters; Performing normalization processing on the normalized operating parameters to obtain normalized operating parameters; A response surface model for each physical field is established according to the reference response value of each physical field under the action of the multiple operating condition sampling points and the normalized operating condition parameters corresponding to the multiple operating condition sampling points.

6. The method according to claim 1, characterized in that The various physical fields include electromagnetic fields, thermal fields, and stress fields.

7. The method according to claim 1, characterized in that Get sample run data and include: Obtain working condition parameters of multiple working condition sampling points; The simulation software is used to output the reference response values ​​of various physical fields of the motor at each working condition sampling point based on the working parameters of each working condition.

8. The method according to claim 1, characterized in that The optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of the individuals in the initial population in each physical field are optimized and calculated to obtain the optimization result, including: A multi-objective optimization algorithm is used to perform optimization calculations based on the optimization function corresponding to the motor optimization objective, the initial population, the initial response values ​​of the individuals in the initial population in each physical field, and the response surface model of at least one physical field, to obtain the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

9. A motor optimization device, characterized in that: The device comprises: a data acquisition module, configured to acquire an initial population, motor optimization targets, and a sample operating data set obtained using simulation software, wherein the sample operating data set includes operating parameters corresponding to a plurality of operating condition sampling points, and reference response values ​​of a plurality of physical fields at each operating condition sampling point output based on the operating parameters of each operating condition during simulation by the simulation software; A model building module, configured to build a response surface model for each physical field based on the operating parameters corresponding to the plurality of operating sampling points and the reference response values ​​of the plurality of physical fields at each operating sampling point; A response value prediction module is used to predict the initial response value of individuals in the initial population in each physical field based on the initial population using the response surface model of each physical field; An optimization calculation module is used to perform optimization calculations based on the optimization function corresponding to the motor optimization target, the initial population, and the initial response values ​​of the individuals in the initial population in each physical field to obtain optimization results, wherein the optimization results include the optimized values ​​of the optimization variables in the optimization function and the optimized response surface model of at least one physical field.

10. An electronic device, characterized in that: include: one or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores program code, and the program code can be called by a processor to execute the method according to any one of claims 1 to 8.

Images