A method for electromagnetic optimal design of a three-phase asynchronous motor and related device

Abstract

This invention belongs to the field of electrical engineering and discloses an electromagnetic optimization design method and related apparatus for a three-phase asynchronous motor. The method includes: acquiring the motor design parameters of the target motor; wherein the motor design parameters of the target motor include basic motor parameters, stator and rotor slot parameters, winding parameters, material cost parameters, and engineering constraints; based on the motor design parameters of the target motor, using a multi-strategy improved whale optimization algorithm to optimize the stator and rotor slot parameters to obtain the optimal stator and rotor slot parameters; and verifying the optimal stator slot parameters based on the two-dimensional transient field-circuit coupled finite element theory to obtain the electromagnetic optimization design result of the target motor. This invention can achieve fast and accurate convergence while ensuring population diversity, effectively avoiding premature entry into local optima, and ultimately obtaining a globally better combination of stator and rotor slot parameters, thereby improving the electromagnetic performance of the motor, reducing material costs, and significantly shortening the design cycle.

Description

Technical Field

[0001] This invention belongs to the field of electrical engineering technology, and relates to motor optimization design technology, particularly to an electromagnetic optimization design method and related apparatus for a three-phase asynchronous motor. Background Technology

[0002] Three-phase asynchronous motors, as AC motors that convert energy based on the principle of electromagnetic induction, have been widely used in industrial production, transportation, new energy power generation, and aerospace. The electromagnetic design of three-phase asynchronous motors is a core element determining the motor's performance, size, cost, and reliability, and its design level directly affects the motor's overall competitiveness. Essentially, the electromagnetic design of three-phase asynchronous motors is a complex optimization problem involving high-dimensionality, strong coupling, and nonlinearity. Traditional electromagnetic design methods mainly rely on engineers' long-term engineering experience, employing trial-and-error and empirical analogy methods, adjusting parameters based on past design cases and their own experience. This approach is inefficient, has a long design cycle, cannot approach the global optimal solution, and struggles to meet the development demands for high-performance, miniaturized, and energy-efficient motors, thus failing to quickly adapt to the customized design requirements of different engineering scenarios.

[0003] Currently, to address the problems in the electromagnetic design of three-phase asynchronous motors, the industry has begun to introduce various intelligent optimization algorithms. Among them, the more widely used intelligent optimization algorithms include Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), which have improved the optimization efficiency and accuracy of motor electromagnetic design to a certain extent and alleviated the challenges brought by high-dimensional optimization problems. In addition, with the rapid development of meta-heuristic algorithms, Whale Optimization Algorithm (WOA), as an emerging meta-heuristic algorithm, has also been gradually applied to the electromagnetic design of three-phase asynchronous motors due to its advantages such as fewer parameter settings, strong local search ability, and fast convergence speed, in an attempt to further improve the optimization effect.

[0004] However, existing standard whale optimization algorithms suffer from several problems when dealing with high-dimensional mixed-integer optimization problems in the electromagnetic design of three-phase asynchronous motors. These problems include uneven initial population distribution, rigid linear decay of the convergence factor, and susceptibility to local optima in the later stages. Specifically, when applied to high-dimensional mixed-integer optimization problems, the standard whale optimization algorithm uses random initialization of the initial population, resulting in uneven population distribution in the parameter space. This fails to fully cover the entire optimization space, affecting the comprehensiveness of the optimization. Secondly, the convergence factor uses a linear decay mode, and this rigid decay method cannot adapt to the complex changes in the optimization space. This leads to slower convergence speed and limited convergence accuracy in the later stages, making it prone to getting trapped in local optima too early and failing to find the globally optimal design parameters. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides an electromagnetic optimization design method and related apparatus for a three-phase asynchronous motor, which solves the problems of uneven initial population distribution, linear decay and rigidity of convergence factor, and easy getting trapped in local optima in the later stage when the existing standard whale optimization algorithm faces the high-dimensional mixed integer optimization problem of electromagnetic design of three-phase asynchronous motors.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: This invention provides an electromagnetic optimization design method for a three-phase asynchronous motor, comprising: Obtain the motor design operating parameters of the target motor; among which, the motor design operating parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints. Based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy. Based on the two-dimensional transient field-circuit coupled finite element theory, the optimal stator slot parameters are verified, and the electromagnetic optimization design results of the target motor are obtained.

[0007] Furthermore, based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm. In the process of obtaining the optimal stator and rotor slot parameters, electromagnetic calculations are performed on the parameters of the current population based on the T-type equivalent circuit of the three-phase asynchronous motor, and the electromagnetic calculation results corresponding to the parameter vector of the current population are obtained. Based on the preset target weights and constraints, and combined with the electromagnetic calculation results corresponding to the parameter vector of the current population, the composite fitness value of the current population is calculated.

[0008] Furthermore, the Circle chaotic mapping sequence is used to generate the initial population; the mathematical model of the Circle chaotic mapping sequence is as follows:

[0009] in, For the first Chaotic state variables generated in the next iteration; For the first Chaotic state variables generated in the next iteration; For modulo operation; , All are predetermined constants; The initial population is obtained by performing a linear mapping process on the chaotic mapping sequence of Circle; the linear mapping process is as follows:

[0010] in, For the first The first individual whale Initial position of the dimension variable; For the first Lower bound of dimensional variable; For the first The upper bound of a dimensional variable.

[0011] Furthermore, the mathematical model representation of the adaptive step size factor is as follows:

[0012]

[0013]

[0014] in, For the first The adaptive step size factor for the next iteration; It is a function for maximizing the value; This is the minimum step size factor; It is a minimum value function; This is the maximum step size factor; For the first The baseline step size factor for the next iteration; It is a decay function; For random disturbance terms; This represents the maximum number of iterations. The attenuation coefficient; This represents the current iteration number.

[0015] Furthermore, based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm. In the process of obtaining the optimal stator and rotor slot parameters, after each iteration, the algorithm is judged to see if it has stalled according to the preset stall judgment condition, and the differential evolution operator is triggered to update the population. The preset stagnation determination condition is as follows: the relative change in the global optimal fitness value within a preset number of consecutive iterations is less than a preset threshold. Differential evolution operators include performing mutation and crossover operations to generate mutation vectors and trial vectors, and updating the population based on mutation vectors and trial vectors, combined with greedy selection and elite retention strategies.

[0016] Furthermore, the mathematical model representation of the oppositional learning strategy is as follows:

[0017] in, For the first The first individual whale The opposite position of the dimensional variable; For the first Lower bound of dimensional variable; For the first Upper bound of a dimensional variable; For the first The first individual whale The current position of the dimension variable.

[0018] Furthermore, the mathematical model representation of the smoothing strategy is as follows:

[0019] in, The position value after smoothing; These are the weighting coefficients; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration.

[0020] This invention also provides an electromagnetic optimization design system for a three-phase asynchronous motor, used to implement the electromagnetic optimization design method for the three-phase asynchronous motor, comprising: The data preprocessing module is used to obtain the motor design operating parameters of the target motor; among which, the motor design operating parameters of the target motor include basic motor parameters, stator and rotor slot parameters, winding parameters, material cost parameters, and engineering constraints. The optimization calculation module is used to optimize the stator and rotor slot parameters based on the motor design operating parameters of the target motor and the multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy. The result verification output module is used to verify the optimal stator slot parameters based on the two-dimensional transient field-circuit coupled finite element theory, and obtain the electromagnetic optimization design results of the target motor.

[0021] The present invention also provides an electronic device, comprising: A processor is used to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, performs the electromagnetic optimization design method for the three-phase asynchronous motor.

[0022] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the electromagnetic optimization design method for the three-phase asynchronous motor.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: The electromagnetic optimization design method for three-phase asynchronous motors provided by this invention utilizes a multi-strategy improved whale optimization algorithm to optimize stator and rotor slot parameters. Based on two-dimensional transient field-circuit coupled finite element theory, the optimal stator slot parameters are verified. This method achieves rapid and accurate convergence while ensuring population diversity, effectively avoiding premature entrapment in local optima, and ultimately obtaining a globally superior combination of stator and rotor slot parameters. This improves the motor's electromagnetic performance, reduces material costs, and significantly shortens the design cycle, meeting the practical engineering requirements for high-performance, miniaturized, energy-efficient, and rapidly customized motor designs. Specifically, by introducing a Circle chaotic mapping sequence for population initialization, the initial solutions are more evenly distributed in the parameter space, effectively improving the population... Diversity avoids the blind spot problem caused by random initialization in standard algorithms; the adaptive step size factor replaces the original linear decay convergence factor, which can dynamically adjust the search step size according to the optimization process, maintaining strong global exploration ability in the early stage of the algorithm and enhancing local fine search ability in the later stage, thus significantly improving convergence accuracy and accelerating convergence speed; the introduction of differential evolution operators enhances the information interaction and mutation ability among populations, effectively escaping the local optimum trap; the opposition learning strategy further expands the search range and improves optimization efficiency by generating opposition solutions and selectively retaining the best ones; the smoothing processing strategy processes discrete variables into continuous variables and then maps them back to integer space, solving the optimization difficulties caused by the coupling of discrete and continuous variables in high-dimensional mixed integer optimization.

[0024] Furthermore, a T-type equivalent circuit is introduced into the optimization iteration process for electromagnetic calculation and a composite fitness evaluation mechanism is constructed. This enables rapid solution of key electromagnetic performance indicators, with a single calculation time only a fraction of that of finite element simulation. Thus, while ensuring the basic accuracy of electromagnetic calculation, the iteration efficiency of each generation of the population is significantly improved, allowing the high-dimensional mixed integer optimization process to complete global optimization within a reasonable time. At the same time, multiple performance indicators are integrated into a composite fitness value by pre-setting target weights, and engineering constraints are embedded into the fitness evaluation using a penalty function mechanism. This allows the algorithm to flexibly adjust the optimization orientation according to the emphasis requirements of different application scenarios and guide the population to converge toward a high-quality solution region that satisfies engineering manufacturability.

[0025] The electromagnetic optimization design system, electronic equipment, and computer-readable storage medium for three-phase asynchronous motors provided by this invention possess all the advantages of the electromagnetic optimization design method for three-phase asynchronous motors described above. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart of the electromagnetic optimization design method for a three-phase asynchronous motor provided by the present invention; Figure 2 A schematic diagram illustrating the principle of the electromagnetic optimization design method for a three-phase asynchronous motor provided in Example 1; Figure 3 This is a structural block diagram of the electromagnetic optimization design system for a three-phase asynchronous motor provided in Example 2; Figure 4 This is a schematic diagram of the structure of the six stator slot types in Example 2; Figure 5 The diagram shows the structure of the four rotor slot types in Example 2; Figure 6 This is a structural block diagram of the electronic device provided in Example 3. Detailed Implementation

[0028] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0029] To address the problems in the existing technology, the inventors discovered in practical work that the whale optimization algorithm simulates the hunting behavior of humpback whales, treating individual whales as search agents and the prey location as the optimal solution. Whales approach the optimal solution through cooperation and information sharing. However, the standard whale optimization algorithm significantly limits the solution quality when facing complex high-dimensional mixed-integer optimization problems. Specifically, the population generated by random initialization is unevenly distributed in the high-dimensional parameter space. If the initial individuals are biased towards inefficient regions, the algorithm will prematurely lock into local suboptimal solutions, making it almost impossible to backtrack to the globally optimal interval later. Secondly, the fixed linear decay of the convergence factor cannot adjust its pace according to the actual search progress. Too rapid decay leads to insufficient exploration in the early stages, while too slow decay results in slow convergence later, both weakening the accuracy of the final solution. As iterations deepen, these defects are further exacerbated in multi-peak problems: when the convergence factor approaches 0, | A The probability of exceeding 1 continues to decrease, and the population as a whole falls into a spiral contraction state. The standard whale optimization algorithm lacks an active mutation perturbation mechanism in the later stages. Once it stagnates at a certain local extreme value, it is difficult to break through autonomously, thus causing premature convergence.

[0030] As attached Figure 1 As shown, this invention provides an electromagnetic optimization design method for a three-phase asynchronous motor, comprising the following steps: Step 100: Obtain the motor design parameters of the target motor; wherein, the motor design parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints.

[0031] Step 200: Based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm incorporates a Circle chaotic mapping sequence, an adaptive step size factor, a differential evolution operator, an opposition learning strategy, and a smoothing processing strategy.

[0032] Step 300: Based on the two-dimensional transient field-circuit coupling finite element theory, the optimal stator slot parameters are verified to obtain the electromagnetic optimization design results of the target motor.

[0033] In the above implementation, the multi-strategy improved whale optimization algorithm, when dealing with the complex mixed-integer optimization problem of high-dimensional, strongly coupled, and nonlinear electromagnetic design of three-phase asynchronous motors, can achieve fast and accurate convergence while ensuring population diversity, effectively avoiding premature entrapment in local optima, and ultimately obtaining a globally better combination of stator and rotor slot parameters, thus effectively improving the optimization efficiency and accuracy of motor electromagnetic design. Specifically, by introducing a Circle chaotic mapping sequence to generate the initial population, the ergodicity and randomness of the Circle chaotic mapping sequence are used to ensure that the population is evenly distributed in the parameter space, fundamentally improving the problem of insufficient population diversity caused by random initialization in the standard whale optimization algorithm, ensuring that the algorithm can fully cover the entire high-dimensional mixed-integer optimization space; Building upon this foundation, an adaptive step-size factor replaces the original linear decay convergence factor, enabling the algorithm to dynamically adjust the step size based on the search state at the current iteration stage. This achieves an adaptive balance between global exploration and local exploitation, effectively avoiding late-stage convergence lag caused by rigid decay patterns. Simultaneously, the embedding of differential evolution operators guides the search direction through inter-individual difference vectors, enhancing the algorithm's ability to handle perturbations in complex nonlinear spaces and escape local optima. An opposition learning strategy is introduced, generating opposition solutions to the current solution and comparing their merits, effectively expanding the search neighborhood and accelerating the convergence process. Furthermore, the introduction of a smoothing strategy constrains parameter jumps during iteration, ensuring the continuity and engineering feasibility of stator and rotor slot parameters within the mixed-integer space.

[0034] The electromagnetic optimization design method for a three-phase asynchronous motor provided by the present invention will be further explained below with some specific embodiments: Example 1 This embodiment 1 provides an electromagnetic optimization design method for a three-phase asynchronous motor. Please refer to the appendix. Figure 2 , attached Figure 2 The diagram below illustrates the principle of the electromagnetic optimization design method for a three-phase asynchronous motor provided in Example 1; specifically, the method includes the following steps: Step 1: Obtain the motor design parameters of the target motor; the motor design parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints.

[0035] Specifically, the basic parameters of the motor include rated power, number of poles, number of stator slots, number of rotor slots, and core length; stator and rotor slot type parameters include the search space of the geometric dimensions of the stator and rotor slots, such as the lower and upper bounds of the search for slot width, slot height, tooth width, and yoke height; winding parameters include wire diameter, number of parallel windings, and number of conductors per slot; material cost parameters include the unit price of silicon steel sheets, the unit price of copper wire, and the unit price of cast aluminum; engineering constraints include the lower limit of power factor, the upper limit of starting current multiple, the range of starting torque multiple, and the upper limit of slot fill factor.

[0036] Optionally, step 1 further includes acquiring a motor simulation model, selecting stator and rotor slot types, configuring winding schemes, and configuring EWOA algorithm parameters; the motor simulation model includes a motor model file established based on magnetic circuit simulation software, or a motor model established based on basic motor parameters; the stator and rotor slot type selection includes several pre-determined stator slot types and pre-determined rotor slot types; the winding scheme configuration includes pre-configured multi-wire diameter hybrid winding schemes; wherein, based on the initial electromagnetic calculation results of each multi-wire diameter hybrid winding scheme, candidate schemes that meet engineering constraints are screened; the EWOA algorithm parameter configuration includes population size, maximum number of iterations, maximum and minimum values ​​of the adaptive step size factor, scaling factor and crossover probability of the differential evolution operator, stagnation detection window, and convergence threshold.

[0037] This embodiment 1 also provides a preferred solution, including: The format and range of the motor design parameters of the target motor are verified; in particular, the motor design parameters of the target motor are verified by verifying the geometric rationality of the air gap length between the stator and rotor.

[0038] Step 2: Based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using the multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters.

[0039] In this embodiment 1, based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm. During the process of obtaining the optimal stator and rotor slot parameters, electromagnetic calculations are performed on the parameters of the current population based on the T-type equivalent circuit of the three-phase asynchronous motor to obtain the electromagnetic calculation results corresponding to the parameter vector of the current population. Based on the preset target weights and constraints, and combined with the electromagnetic calculation results corresponding to the parameter vector of the current population, the composite fitness value of the current population is calculated.

[0040] Specifically, the multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces a Circle chaotic mapping sequence, an adaptive step size factor, a differential evolution operator, an opposition learning strategy, and a smoothing strategy. The specific implementation process of introducing the Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy, and smoothing strategy is as follows: (1) Circle chaotic mapping sequence Explained, in intelligent optimization algorithms, the distribution quality of the initial population has a significant impact on the algorithm's search performance. The standard whale optimization algorithm usually uses a uniform random method to generate population individuals in the initialization phase. However, random initialization may lead to uneven distribution of individuals in the search space, resulting in insufficient search in some areas, reducing the algorithm's global search capability, and potentially causing the algorithm to get stuck in local optima in the early iteration phase.

[0041] In this embodiment 1, to improve the quality of the initial population distribution, a Circle chaotic mapping sequence is introduced to generate the initial population. The Circle chaotic mapping sequence possesses ergodicity, quasi-randomness, and sensitivity to initial conditions, enabling it to form a more uniform distribution within the search space, thereby increasing population diversity and enhancing the algorithm's global search capability. Furthermore, the Circle chaotic mapping sequence has a simple structure and good ergodicity. The mathematical model representation of the Circle chaotic mapping sequence is as follows:

[0042] in, For the first Chaotic state variables generated in the next iteration; For the first Chaotic state variables generated in the next iteration; This is a modulo operation used to ensure that the sequence always falls within the interval [0,1]. , All are predetermined constants; preferably, The value is 0.2. The value is 0.5; it is worth noting that the chaotic state variable generated in the 0th iteration... The initial value is a random value in the interval [0,1].

[0043] Using the mapping relationship represented by the mathematical model of the Circle chaotic mapping sequence described above, it is possible to generate Circle chaotic mapping sequences with good ergodicity. In practical applications, in order to map the Chaotic Circle sequence... Mapping to the variable space of the optimization problem requires linear mapping of the chaotic sequence, specifically linear mapping of the Circle chaotic mapping sequence, to obtain the initial population. An example of the linear mapping process is illustrated below: Assume the optimization problem is of the th order. The range of values ​​for each variable is Then the first i The individual in the first j The initial position of the dimension can be represented as:

[0044] Among them, among them, For the first The first individual whale Initial position of the dimension variable; For the first Lower bound of dimensional variable; For the first The upper bound of a dimensional variable.

[0045] In this embodiment 1, the initial population generated by linearly mapping the Circle chaotic mapping sequence helps to form a more dispersed and better-covered initial population in the search space, thereby improving population diversity and enhancing the algorithm's early global search capability, and improving the algorithm's convergence speed and optimization accuracy.

[0046] (2) Adaptive step size factor Explained, in the standard whale optimization algorithm, individual position updates are primarily determined by control parameters. A , C and random variables p Jointly decided; when satisfied p <0.5 and | A When |≥1, individuals update their positions by referencing random individuals, corresponding to the global search phase of the algorithm. This phase undertakes the task of searching a large area of ​​the population and is quite sensitive to changes in the search step size. As the iteration progresses, if the global search step size lacks a dynamic adjustment mechanism, it is easy to encounter problems such as insufficient search range in the early stage or excessive random perturbation in the later stage, thereby affecting the search efficiency and convergence stability of the algorithm.

[0047] Therefore, in this embodiment 1, an adaptive step size factor is introduced in the global search phase. By dynamically scaling the displacement terms of the random search branches, the algorithm maintains a strong global exploration capability in the early stages of iteration and gradually reduces the random jump amplitude in the later stages of iteration, thereby achieving adaptive adjustment of the search intensity. Specifically, to ensure that the search step size gradually decreases during the iteration process, a baseline step size factor is defined as follows:

[0048] in, The minimum step size factor is set to 0.4. The maximum step size factor is set to 0.9; For the first The baseline step size factor for the next iteration; This represents the maximum number of iterations.

[0049] As can be seen from the definition of the baseline step size factor, it decreases linearly with the number of iterations. In the early stages of iteration, its value is relatively large, which helps to expand the step size range of the random search phase and enhance the population's coverage of the search space. In the later stages of iteration, its value gradually decreases, which helps to suppress invalid random jumps and improve the convergence stability of the algorithm. To further enhance the flexibility of the search process, a random perturbation term is introduced based on the baseline step size factor, and the perturbation amplitude is controlled by a decay function. The decay function is defined as follows:

[0050] in, It is a decay function; This represents the maximum number of iterations. The attenuation coefficient; This represents the current iteration number.

[0051] Given the defined baseline step size factor and decay function, the mathematical model for the adaptive step size factor is as follows:

[0052] in, For the first The adaptive step size factor for the next iteration; It is a function for maximizing the value; This is the minimum step size factor; It is a minimum value function; This is the maximum step size factor; For the first The baseline step size factor for the next iteration; It is a decay function; This is a random disturbance term, generated by a random variable, with a disturbance range of (-1, 1); This represents the maximum number of iterations. The attenuation coefficient; This represents the current iteration number.

[0053] It should be noted that the adaptive step size factor maintains a large value in the early stage of iteration, accompanied by a certain degree of random fluctuation, which is beneficial to enhance the global search capability of the population. As the number of iterations increases, the adaptive step size factor generally shows a downward trend, and the fluctuation amplitude gradually shrinks, making the random search branch more stable in the later stage, thereby improving the search efficiency and convergence stability of the algorithm.

[0054] (3) Differential evolution operator Interpretive, Differential Evolution (DE) is a classic evolutionary computation technique. Proposed by Storm et al. in 1995 to solve the Chebyshev polynomial problem, subsequent research has shown that DE also exhibits strong global search capabilities in handling complex continuous optimization problems, becoming one of the important methods in the field of evolutionary computation. Differential Evolution is an optimization algorithm based on swarm intelligence theory. It uses a global search strategy generated through cooperation and competition among individuals within the swarm, employing real number encoding, simple mutation operations based on differences, and a one-to-one competitive survival strategy to reduce the complexity of evolutionary computation operations.

[0055] In this embodiment 1, the differential evolution operator is embedded in the iterative update process of the whale optimization algorithm to enhance population diversity and improve the ability to escape local optima when the population search stagnates. After each iteration, a preset stagnation determination condition is used to determine whether the algorithm has stagnated, triggering the differential evolution operator to update the population. Preferably, the preset stagnation determination condition is: the relative change in the global optimal fitness value within a consecutive preset number of iterations is less than a preset threshold. That is, considering that performing the DE operation in every iteration might interfere with the normal search rhythm of the WOA, this embodiment 1 uses a stagnation detection trigger mechanism to control the execution timing of the DE operator. Specifically, when consecutive iterations... The relative change in the global optimal fitness value within a generation is less than the threshold. When the algorithm stalls, the differential evolution operator is triggered to begin its evolutionary process. The specific conditions for this are as follows:

[0056] in, For the first Replace the globally optimal fitness value; To prevent extremely small constants with a denominator of zero, we take 10. 10 ; For the stall detection window, set the value to 10; The threshold value is set to 10. 6 .

[0057] In this embodiment 1, the differential evolution operator includes performing mutation and crossover operations to generate mutation vectors and trial vectors. Based on the mutation vectors and trial vectors, the population is updated using a greedy selection strategy and an elite retention strategy. Specifically, in each iteration, all individuals are first updated according to the WOA position update mechanism. The current population after the WOA update is denoted as... , indicating the first iThe current position of each individual is updated by WOA; then the defined stagnation condition is checked; if the stagnation condition is not met, the current population is directly set to [position not specified]. As the next generation population; if the stagnation condition is met, then the mutation, crossover and selection operations of DE are performed on the basis of the current population; preferably, in order to avoid disturbing and destroying the current best individual, this embodiment 1 adopts an elite retention strategy, that is, the current global best individual is directly retained and does not participate in the replacement operation.

[0058] The mutation operation is the core step of the DE algorithm, used to generate mutation vectors using the difference information of individuals within the population, thereby enhancing population diversity; for the ... For each individual, the mutation operation generates a mutation vector using the following formula. .

[0059]

[0060] in, , , For three distinct indices of randomly selected individuals in the population, none of them are equal to 1. ; This is the scaling factor, used to control the scaling degree of the difference vector.

[0061] The crossover operation will mutate the vector With the target vector Exchange information and generate test vectors Among them, the test vector The The definition of a dimensional component is as follows:

[0062] in, For the test vector The j dimensional components, , From the perspective of the problem dimension; for; The index is randomly selected from [1, D].

[0063] To determine whether the experimental vector is better than the current target vector, this embodiment 1 employs a greedy selection strategy; where, for the minimization problem, the... i The update rules for each individual are as follows:

[0064] in, For the updated number i The location of each individual.

[0065] It should be noted that, considering the crucial guiding role of the current optimal individual in population convergence, an elite retention strategy is adopted in Implementation Example 1 to avoid DE perturbation from disrupting the existing optimal solution. Specifically, after the WOA completes the position update, the optimal individual in the current population is first identified and directly retained, without participating in subsequent DE mutation, crossover, and selection replacement operations; the remaining individuals complete the evolution update according to the DE rules. This strategy can enhance population diversity while ensuring that the current global optimal solution is not destroyed.

[0066] (4) Oppositional learning strategies Interpretive, contrastive learning strategies can effectively expand the search scope of a population, uncover new search regions, and enhance population diversity. Combining these strategies with swarm intelligence optimization algorithms can improve the algorithm's global search capability and prevent premature convergence. In this embodiment 1, a contrastive learning strategy is used to enhance the global search capability. Specifically, by generating contrastive solutions to the current solution, the search space is expanded, population diversity is enhanced, and the algorithm's global search capability is improved. The specific expression for the contrastive learning strategy is as follows:

[0067] in, For the first The first individual whale The opposite position of the dimensional variable; For the first Lower bound of dimensional variable; For the first Upper bound of a dimensional variable; For the first The first individual whale The current position of the dimension variable.

[0068] It is worth noting that, in order to combine the opposition learning strategy with the whale optimization algorithm, this embodiment 1 introduces an opposition solution generation mechanism during the population update process to improve population diversity and enhance global search capabilities; firstly, in the population initialization phase, an initial population is randomly generated. X The original population is calculated based on the opposition learning strategy, and then the opposition population is merged with the original population. The fitness values ​​of all individuals are calculated, and the individuals with better fitness are selected. N Individuals form a new initial population, thereby improving population quality and expanding the search area. During the algorithm iteration process, after an individual completes its position update, it also uses the opposite learning strategy to generate its opposite solution and compares the fitness values ​​of the two. If the fitness of the opposite solution is better than that of the current individual, the original individual is replaced by the opposite solution, thereby enhancing the search diversity of the population and improving its ability to escape local optima. By introducing the opposite learning strategy, the algorithm can simultaneously consider the current solution and its opposite solution in the search space, expand the search range, improve the global search capability, and thus improve the overall optimization performance of the improved WOA algorithm.

[0069] (5) Smoothing strategy In this embodiment 1, by introducing the Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, and opposition learning strategy, the initialization and iterative search phases of the standard whale optimization algorithm can be improved. A smoothing strategy is introduced to address the oscillation problem in the later stages of convergence. Specifically, during the search process of the WOA algorithm, especially in the later stages of iteration, the search trajectory of the current optimal individual may exhibit local or short-term oscillations due to the influence of multi-strategy superposition perturbation and random update mechanism, thereby reducing the convergence stability of the algorithm and affecting the final solution accuracy.

[0070] In this embodiment 1, in order to suppress abnormal fluctuations in the later search process, an embedded smoothing processing strategy is introduced. While maintaining the algorithm's search capability, the update trajectory of the current best individual in the later stage is locally smoothed to improve the stability of the algorithm's convergence phase.

[0071] Specifically, the smoothing strategy is directly embedded into the optimization iteration process and applied to the position sequence of the current best individual in the later stages of convergence; let... Represents the first dimension k In the next iteration, when the position value of the current best individual is reversed between two adjacent segments, it can be considered that the position sequence has experienced local oscillations near that point. In this case, a smoothing strategy is used for correction. The mathematical model of the smoothing strategy is as follows:

[0072] in, The position value after smoothing; These are the weighting coefficients; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration.

[0073] It should be noted that in the multi-strategy improved WOA algorithm, the smoothing strategy is only activated in the later stages of the iteration to correct local oscillations in various dimensions of the current best individual. In this way, abnormal jumps in the later stages of convergence can be reduced without significantly weakening the algorithm's search capability, thereby enhancing the algorithm's stability and improving the final solution accuracy.

[0074] In this embodiment 1, by introducing the Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy, and smoothing strategy, a systematic enhancement of the standard WOA is achieved from five dimensions: initialization, global exploration, population diversity, local exploitation, and convergence stability. Specifically, the Circle chaotic mapping ensures uniform coverage of the initial population; the adaptive step size factor dynamically balances global exploration in the early stages of iteration with fine exploitation in the later stages; the differential evolution operator injects new mutation momentum when the population stagnates, maintaining population diversity; the opposition learning strategy expands the search space and enhances global exploration capabilities; and the smoothing mechanism eliminates oscillations in the later stages of convergence, ensuring the stability of the solution. These five strategies work together to form an improved algorithm system with both powerful global search capabilities and efficient local exploitation capabilities, providing reliable algorithmic support for the subsequent efficient and low-cost optimization design of three-phase asynchronous motors.

[0075] Specifically, the process of optimizing the stator and rotor slot parameters based on the motor design operating parameters of the target motor, using a multi-strategy improved whale optimization algorithm, is as follows: S1. Initialization Phase: An initial population is generated using a Circle chaotic mapping sequence. The chaotic sequence values ​​are mapped to the variable space of the optimization problem through a linear mapping to obtain the initial whale individual positions. An opposing population is generated based on an opposing learning strategy. The original population and the opposing population are merged, and the top N individuals are selected according to their fitness values ​​to form a new initial population.

[0076] S2, Optimization Iteration Phase: A composite fitness function is constructed based on motor efficiency and cost. Electromagnetic calculations are performed based on a T-type equivalent circuit to obtain efficiency, cost, power factor, starting current multiple, and starting torque multiple. Specifically, based on the T-type equivalent circuit of a three-phase asynchronous motor, stator winding parameters, rotor equivalent parameters, and excitation parameters are calculated. On this basis, each loss component and efficiency are calculated. At the same time, power factor, starting torque, and starting current are calculated. This mode has a short single calculation time and is suitable for large-scale iterative calls.

[0077] The adaptive step size factor is calculated based on the current iteration number; the adaptive step size factor is composed of a baseline step size factor that decreases linearly with the iteration number and a random perturbation term controlled by a decay function; the position is updated according to the three mechanisms of the whale optimization algorithm: surrounding prey, spiral bubble web attack, and search and predation.

[0078] After each iteration, check for stall conditions; where, when continuous When the relative change in the global optimal fitness value within a generation is less than a preset threshold, the algorithm is deemed to have stalled, triggering the differential evolution operator. The differential evolution operator performs mutation and crossover operations to generate test vectors, and uses a greedy selection strategy and an elite retention strategy to update the population.

[0079] In the later stages of iteration, the position sequence of each dimension of the current best individual is smoothed; specifically, when the direction of change of adjacent positions reverses, a weighted average method is used to correct the intermediate position value and suppress local oscillations.

[0080] S3, Termination Judgment Phase: Determine whether the maximum number of iterations has been reached or the convergence condition has been met; if not, return to step S2 to continue iterating; if yes, output the optimal stator and rotor slot parameters.

[0081] Step 3: Based on the two-dimensional transient field-circuit coupled finite element theory, the optimal stator slot parameters are verified to obtain the electromagnetic optimization design results of the target motor.

[0082] The specific implementation process is as follows: Step 31: Import the optimal stator and rotor slot parameters obtained in Step 2 into the pre-acquired motor simulation model, configure the simulation boundary conditions according to the rated operating conditions of the target motor, and obtain the finite element simulation model of the target motor under rated operating conditions; wherein, the rated operating conditions of the target motor include rated voltage, rated frequency and rated load torque; configure the simulation boundary conditions based on the rated operating conditions of the target motor to ensure that the simulation operating conditions are consistent with the actual operating conditions, thereby ensuring the reliability of the verification results.

[0083] Step 32: Based on the finite element simulation model of the target motor under rated operating conditions, run the transient electromagnetic field simulation in the pre-determined finite element simulation software to obtain the magnetic flux density distribution cloud map and transient torque-speed curve of the target motor under rated operating conditions.

[0084] Step 33: Based on the magnetic flux density distribution cloud map of the target motor under rated operating conditions, extract the peak magnetic flux density of preset key parts in the target motor; the preset key parts include, for example, stator teeth, stator yoke, rotor teeth, and rotor yoke; substitute the extracted peak magnetic flux density of the preset key parts in the target motor into the predetermined magnetic circuit saturation judgment equation to obtain the magnetic saturation simulation value of the preset key parts; compare the magnetic saturation simulation value of the preset key parts with the preset magnetic saturation judgment threshold; if the magnetic saturation simulation value of all preset key parts does not exceed the preset magnetic saturation judgment threshold, the magnetic circuit design of the target motor is determined to be reasonable, and the magnetic flux density verification passes; if the magnetic saturation simulation value of a certain preset key part exceeds the preset magnetic saturation judgment threshold, the optimal stator and rotor slot parameters are determined to cause local magnetic circuit saturation in the target motor, and the magnetic flux density verification fails.

[0085] Step 34: Based on the transient torque-speed curve of the target motor under rated operating conditions, extract the starting torque multiple, maximum torque multiple, and rated speed in sequence. Compare the extracted starting torque multiple, maximum torque multiple, and rated speed with the preset starting torque multiple range, maximum torque multiple lower limit, and rated speed allowable deviation, respectively. If all performance indicators meet the preset engineering constraints, the electromagnetic performance verification is deemed to have passed. If any performance indicator does not meet the preset engineering constraints, the electromagnetic performance verification is deemed to have failed.

[0086] Step 35: If both the magnetic flux density verification and electromagnetic performance verification pass, the optimal stator and rotor slot parameters are output as the electromagnetic optimization design result of the target motor. If either verification fails, the constraint information that does not meet the conditions is used as feedback to update the corresponding constraint boundary in the engineering constraints. Then, return to step 2 and re-execute the multi-strategy improved whale optimization algorithm for optimization iteration until the verification passes. The electromagnetic optimization design result of the target motor includes the optimal stator and rotor slot parameters, performance index comparison, loss distribution analysis, and a structured report of constraint verification results.

[0087] Verification instructions: (1) The performance of the multi-strategy improved whale optimization algorithm was verified on the IEEE CEC standard test set. The verification results showed that the multi-strategy improved whale optimization algorithm could achieve the best or tied best results on 12 IEEE CEC standard test functions. Its convergence accuracy and robustness were significantly better than the mainstream algorithms such as standard whale optimization algorithm, particle swarm optimization algorithm, genetic algorithm and gray wolf optimization algorithm. It effectively solved the problems of insufficient population diversity and local premature convergence of standard whale optimization algorithm in the high-sensitivity space of motor optimization.

[0088] (2) Taking a 7.5kW three-phase asynchronous motor as an example; the multi-strategy improved whale optimization algorithm can increase the motor efficiency from 90.43% to 91.95%, reduce the total electromagnetic cost by 4.84%, and its comprehensive performance is significantly better than the three comparison algorithms WOA, MA and ABC. The simulation error of key parameters is controlled within 5% in the prototype test.

[0089] The electromagnetic optimization design method for the three-phase asynchronous motor described in Embodiment 1 introduces a Circle chaotic mapping sequence for population initialization, making the initial solution more evenly distributed in the parameter space, effectively improving population diversity and avoiding the search blind zone problem caused by random initialization in the standard algorithm. An adaptive step size factor replaces the original linear decay convergence factor, dynamically adjusting the search step size according to the optimization process. This maintains strong global exploration capabilities in the early stages of the algorithm and enhances local fine-grained search capabilities in the later stages, significantly improving convergence accuracy and speed. The introduction of differential evolution operators enhances information interaction and mutation capabilities among the population, effectively escaping local optima traps. The opposition learning strategy generates and selectively retains opposition solutions, further expanding the search range and improving optimization efficiency. The smoothing strategy processes discrete variables into continuous variables before mapping them back to the integer space, solving the optimization difficulties caused by the coupling of discrete and continuous variables in high-dimensional mixed integer optimization.

[0090] Example 2 As attached Figure 3 As shown, this embodiment 2 provides an electromagnetic optimization design system for a three-phase asynchronous motor, used to implement the electromagnetic optimization design method for a three-phase asynchronous motor described in embodiment 1 above, including a data preprocessing module, an optimization calculation module, and a result verification output module.

[0091] The data preprocessing module is used to obtain the motor design operating parameters of the target motor. The motor design operating parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints.

[0092] The optimization calculation module is used to optimize the stator and rotor slot parameters based on the motor design operating parameters of the target motor and using a multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy.

[0093] The result verification output module is used to verify the optimal stator slot parameters based on the two-dimensional transient field-circuit coupled finite element theory, and obtain the electromagnetic optimization design results of the target motor.

[0094] Optionally, the electromagnetic optimization design system for the three-phase asynchronous motor described in Embodiment 2 is developed using Python and the PyQt5 framework. It uses a multi-strategy improved whale optimization algorithm as an embedded optimization engine and integrates an electromagnetic calculation kernel to achieve full-process automation from parameter input, electromagnetic simulation, intelligent optimization to result management.

[0095] In the data preprocessing module, the motor simulation model is selected through a preset interface, the stator and rotor slot types are selected and parameters are configured, material cost parameters and engineering constraints are set, and EWOA algorithm parameters are configured. After the input data is automatically verified for rationality, the magnetic circuit simulation module is called to complete the initial motor performance calculation and generate optimized benchmark data.

[0096] Specifically, the data preprocessing module is the system's input layer, responsible for motor parameter configuration and model initialization. It includes the following modules: simulation model management module, stator and rotor slot type visualization selection module, parameter input and verification module, winding scheme configuration module, and EWOA algorithm parameter configuration module.

[0097] The simulation model management module supports importing motor model files created using magnetic circuit simulation software, and automatically parses and loads the basic parameter information of the model; at the same time, it supports users to directly input parameters to call the electromagnetic calculation module for rapid modeling; the two modeling modes can be switched as needed.

[0098] The stator and rotor slot type visualization selection module provides a graphical selection interface for six stator slot types (S1~S6) and four rotor slot types (R1~R4), as shown in the attached diagram. Figure 4-5 As shown; each slot type is accompanied by a dimensioned schematic diagram; after the user selects a slot type, the system dynamically generates the corresponding parameter input form and automatically verifies the geometric rationality of the stator and rotor air gap length.

[0099] The parameter input and verification module is used to receive user input of basic motor parameters, winding parameters, stator and rotor slot dimensions, material cost parameters, and engineering constraints; it is also used to perform real-time format and range verification on the input data, and abnormal parameters are highlighted.

[0100] The winding scheme configuration module supports configuring multi-wire diameter mixed winding schemes and automatically filters candidate schemes that meet the basic performance thresholds for users to choose from based on the initial electromagnetic calculation results of each winding scheme.

[0101] The EWOA algorithm parameter configuration module is used to set the population size, maximum number of iterations, maximum and minimum values ​​of the adaptive step size factor, scaling factor and crossover probability of the differential evolution operator, stagnation detection window and convergence threshold; the system provides preset parameter templates for different motor types.

[0102] In the optimization calculation module, the EWOA algorithm is used as the optimization engine. The Python main control program realizes a fully automatic closed-loop call between the algorithm and the electromagnetic calculation module: In each iteration, EWOA first generates candidate parameter combinations, then the magnetic circuit simulation module completes the electromagnetic calculation and returns the efficiency, cost and various constraint indicators; then, the fitness evaluation module calculates the composite fitness value; then, EWOA updates the population position based on the evaluation results; the iteration continues until the termination condition is met; during the optimization process, the progress window displays the iteration number, the current optimal efficiency, the optimal cost and the algorithm log in real time.

[0103] Specifically, the optimization calculation module is the core processing layer of the system, which includes the EWOA optimization driving module, the electromagnetic calculation module, and the fitness evaluation module.

[0104] The EWOA optimization driver module is used to implement the multi-strategy improved whale optimization algorithm. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that integrates Circle chaotic mapping initialization, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy to drive efficient search of parameter space.

[0105] The electromagnetic calculation module is used to batch input the parameter vectors of the current population into the adapter in each iteration. The adapter calls the magnetic circuit simulation module to complete the electromagnetic field calculation of all individuals and returns the efficiency, cost, power factor, starting current multiple, and starting torque multiple to the fitness evaluator for calculation. The magnetic circuit simulation module is used to calculate the stator and rotor copper loss, core loss, and mechanical additional loss based on the T-type equivalent circuit to complete the full parameter evaluation of efficiency and starting characteristics.

[0106] The fitness evaluator module is used to construct a composite fitness function based on motor efficiency and cost according to the target weights and constraints set by the user, and to calculate the composite fitness value of each parameter vector. Among them, the penalty term in the composite fitness function constructed by motor efficiency and cost adopts a dynamic weight coefficient, which automatically adjusts the constraint penalty intensity as the iteration process, ensuring the balance of the search for feasible solutions.

[0107] The real-time monitoring interface module is used to display the computation progress, current iteration number, current optimal fitness value, corresponding efficiency and cost values, and constraint satisfaction status in real time through the progress window during the optimization process; the computation log records key node events in timestamp format and supports pausing and resuming midway.

[0108] The electromagnetic calculation kernel is used to invoke different precision calculation modes during the optimization iteration phase and the result verification phase. The electromagnetic calculation kernel includes a magnetic circuit calculation mode and a field-circuit coupled finite element precise verification mode. Specifically, the magnetic circuit calculation mode is used in the optimization iteration phase. The process includes: calculating stator winding parameters, rotor equivalent parameters, and excitation parameters based on the T-type equivalent circuit of a three-phase asynchronous motor; calculating various loss components and efficiency based on these; and simultaneously calculating the power factor, starting torque, and starting current. This mode has a short single calculation time and is suitable for large-scale iteration. The field-circuit coupled finite element precise verification mode is used in the result verification phase. The process includes: solving the electromagnetic field control equations, stator external circuit equations, and rotor squirrel-cage circuit equations simultaneously based on two-dimensional transient field-circuit coupled finite element theory. This yields precise results such as magnetic flux density distribution cloud maps and transient torque-speed curves, used for the final verification of the optimized scheme.

[0109] In this embodiment 2, the result verification output module verifies the optimal stator slot parameters based on the two-dimensional transient field-circuit coupled finite element theory to obtain the electromagnetic optimization design results of the target motor. The electromagnetic optimization design results of the target motor are output in the form of a structured result report, which is used to display the optimal design parameters, performance index comparison, loss distribution analysis and constraint condition verification results. The results can be exported as an Excel file or stored in a database.

[0110] Optionally, this embodiment 2 also includes a result visualization and data management module; the result visualization and data management module is used for the structured display of optimization results, performance curve visualization, database management, and Excel report export; specifically, the result visualization and data management module includes a structured display module for optimization results, a performance curve visualization module, a database management module, and an Excel report export module.

[0111] The optimization results structured display module automatically pops up a results window after optimization, displaying four sections: an optimization summary section, a main performance index comparison section, a loss distribution analysis section, and an optimal design parameter section. Specifically, the optimization summary section displays information including the algorithm name, iteration number, optimal fitness, and whether all constraints are met; the main performance index comparison section displays information including the comparison of rated efficiency, power factor, starting current multiple, starting torque multiple, slot fill factor, and cost reduction rate before and after optimization; the loss distribution analysis section displays information including the proportions and changes of stator copper loss, rotor copper loss, core loss, and stray loss before and after optimization; and the optimal design parameter section displays information including the optimized dimensions of each slot type, winding scheme, and constraint verification results.

[0112] The performance curve visualization module is used to generate EWOA optimization convergence curves, i.e., efficiency and cost changes with iteration number, multi-algorithm comparison convergence curves, as well as motor efficiency-load characteristic curves and power factor-load characteristic curves. The curves can be scaled and saved as images.

[0113] The database management module is used to build three types of databases and to query, compare, and delete historical records in the database interface. It also supports one-click loading of historical parameters from the product database or temporary database to the input interface. The three types of databases include a temporary database, a product database, and a silicon steel sheet material database. The temporary database is used to store the calculation process data of the current session. The product database is used to store the verified optimization schemes, which include complete parameters and performance indicators. The silicon steel sheet material database is used to store the BH curves and loss coefficient material parameters of different grades of silicon steel sheets.

[0114] The Excel report export module supports exporting optimization results to Excel format files, including a summary table of input parameters, an optimization iteration history table, and a detailed parameter table of the optimal solution, which facilitates users in archiving and secondary analysis.

[0115] Example 3 As attached Figure 6 As shown, this embodiment 3 provides an electronic device, including: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the electromagnetic optimization design method for a three-phase asynchronous motor; or, the processor for executing the computer program to implement the functions of each module in the electromagnetic optimization design system for the three-phase asynchronous motor.

[0116] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a preset function, the instruction segments describing the execution process of the computer program in the electronic device.

[0117] The electronic device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are examples of electronic devices and do not constitute a limitation on the electronic device. It may include more components than described above, or combine certain components, or different components. For example, the electronic device may also include a communication interface, input / output devices, network access devices, and a bus.

[0118] The processor can be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (OPGs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor, or any conventional processor. The processor is the control center of the electronic device, connecting various parts of the electronic device via various communication interfaces and lines.

[0119] The memory can be used to store the computer program and / or module. The processor implements various functions of the electronic device by running or executing the computer program and / or module stored in the memory and by calling the data stored in the memory.

[0120] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function (such as sound playback, image playback, etc.). The data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart memory cards, secure digital cards, flash memory cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0121] Example 4 This embodiment 4 also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the electromagnetic optimization design method for a three-phase asynchronous motor.

[0122] If the modules / units integrated in the electromagnetic optimization design system of the three-phase asynchronous motor are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0123] Based on this understanding, the electromagnetic optimization design method for the three-phase asynchronous motor described above can be implemented, either in whole or in part, by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the electromagnetic optimization design method for the three-phase asynchronous motor. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form.

[0124] The computer-readable storage medium may include any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0125] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

Claims

1. An electromagnetic optimization design method for a three-phase asynchronous motor, characterized in that, include: Obtain the motor design operating parameters of the target motor; among which, the motor design operating parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints. Based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy. Based on the two-dimensional transient field-circuit coupled finite element theory, the optimal stator slot parameters are verified, and the electromagnetic optimization design results of the target motor are obtained.

2. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, Based on the motor design parameters of the target motor, a multi-strategy improved whale optimization algorithm is used to optimize the stator and rotor slot parameters. In the process of obtaining the optimal stator and rotor slot parameters, electromagnetic calculations are performed on the parameters of the current population based on the T-type equivalent circuit of the three-phase asynchronous motor to obtain the electromagnetic calculation results corresponding to the parameter vector of the current population. Based on the preset target weights and constraints, and combined with the electromagnetic calculation results corresponding to the parameter vector of the current population, the composite fitness value of the current population is calculated.

3. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, The Circle chaotic mapping sequence is used to generate the initial population; the mathematical model of the Circle chaotic mapping sequence is as follows: in, For the first Chaotic state variables generated in the next iteration; For the first Chaotic state variables generated in the next iteration; For modulo operation; , All are predetermined constants; The initial population is obtained by performing a linear mapping process on the chaotic mapping sequence of Circle; the linear mapping process is as follows: in, For the first The first individual whale Initial position of the dimension variable; For the first Lower bound of dimensional variable; For the first The upper bound of a dimensional variable.

4. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, The mathematical model representation of the adaptive step size factor is as follows: in, For the first The adaptive step size factor for the next iteration; It is a function for maximizing the value; This is the minimum step size factor; It is a minimum value function; This is the maximum step size factor; For the first The baseline step size factor for the next iteration; It is a decay function; For random disturbance terms; This represents the maximum number of iterations. The attenuation coefficient; This represents the current iteration number.

5. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, Based on the motor design operating parameters of the target motor, the stator and rotor slot parameters are optimized using a multi-strategy improved whale optimization algorithm. In the process of obtaining the optimal stator and rotor slot parameters, after each iteration, the algorithm is judged to be stuck according to the preset stagnation judgment condition, and the differential evolution operator is triggered to update the population. The preset stagnation determination condition is as follows: the relative change in the global optimal fitness value within a preset number of consecutive iterations is less than a preset threshold. Differential evolution operators include performing mutation and crossover operations to generate mutation vectors and trial vectors, and updating the population based on mutation vectors and trial vectors, combined with greedy selection and elite retention strategies.

6. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, The mathematical model representation of the oppositional learning strategy is as follows: in, For the first The first individual whale The opposite position of the dimensional variable; For the first Lower bound of dimensional variable; For the first Upper bound of a dimensional variable; For the first The first individual whale The current position of the dimension variable.

7. The electromagnetic optimization design method for a three-phase asynchronous motor according to claim 1, characterized in that, The mathematical model representation of the smoothing strategy is as follows: in, The position value after smoothing; These are the weighting coefficients; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration; For a certain dimension, the first The position value of the current best individual at the next iteration.

8. An electromagnetic optimization design system for a three-phase asynchronous motor, characterized in that, The electromagnetic optimization design method for implementing the three-phase asynchronous motor as described in any one of claims 1-7 includes: The data preprocessing module is used to obtain the motor design operating parameters of the target motor; among which, the motor design operating parameters of the target motor include the basic parameters of the motor, the stator and rotor slot parameters, the winding parameters, the material cost parameters, and the engineering constraints. The optimization calculation module is used to optimize the stator and rotor slot parameters based on the motor design operating parameters of the target motor and the multi-strategy improved whale optimization algorithm to obtain the optimal stator and rotor slot parameters. The multi-strategy improved whale optimization algorithm is a whale optimization algorithm that introduces Circle chaotic mapping sequence, adaptive step size factor, differential evolution operator, opposition learning strategy and smoothing processing strategy. The result verification output module is used to verify the optimal stator slot parameters based on the two-dimensional transient field-circuit coupled finite element theory, and obtain the electromagnetic optimization design results of the target motor.

9. An electronic device, characterized in that, include: A processor is used to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, performs the electromagnetic optimization design method for a three-phase asynchronous motor as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the electromagnetic optimization design method for a three-phase asynchronous motor as described in any one of claims 1-7.

Images