CN115630556A - A motor topology optimization method based on vertex method - Google Patents

Abstract

The invention discloses a motor topology optimization method based on a vertex method, which comprises the following steps: s1, dividing a design area of a motor into m parts, and taking a vertex position as a variable; and S2, connecting all vertexes in the same region into a polygon, taking the overlapped part of the polygon and the corresponding sub-region as the shape of one material in the sub-region, and taking the rest part in the sub-region as the other material. Determining a motor topological structure according to the distribution of materials in each subarea; s3, selecting a target to be optimized according to design requirements, and optimizing coordinate parameters of the vertex by using an intelligent optimization algorithm to obtain a series of optimal vertex positions which enable the motor performance to meet the requirements; s4, judging whether the performance of the motor meets the design requirement, and if so, finishing the design; otherwise, the step S1 is returned to. The invention can realize the automatic design of the high-performance motor on the premise of reducing human intervention.

Description

A motor topology optimization method based on vertex method

technical field

The invention relates to the technical field of motors, in particular to a design method for topology optimization of motors.

Background technique

With the continuous innovation and rapid development of modern industry, motors, as commonly used electromechanical energy conversion equipment, are distributed in various fields of the national economy. With the continuous improvement of motor performance requirements, the previous topology is difficult to meet the requirements. Therefore, innovations in topology are urgently needed to meet the increasing performance requirements of motors.

At present, the optimization of motor topology is mainly to parameterize some new topological structures, and then use intelligent algorithms to optimize them. There are three main problems with such an approach: First, it is extremely demanding for researchers to study new topologies, and it takes a lot of time to perform experimental simulations. Secondly, although the new topology can meet the performance requirements of the motor, its structure is very complex, and it is difficult to abstract parameters to represent its topology. Finally, even if some parameters can be used to represent its shape, the optimized result is limited by the initial structure, and it is difficult to generate new structural shapes. This means that it is difficult to jump out of the local optimal solution and enter the global optimal solution during the optimization process.

In summary, the existing technology fails to achieve the goal of designing a high-performance motor without relying on the existing structure. This patent proposes a method that uses vertices to represent the shape of the motor, and optimizes the position of the vertices through an intelligent algorithm to realize the free evolution of the topology. method.

Contents of the invention

Technical problem: The purpose of the invention is to address the shortcomings of the above-mentioned background technology and provide a motor topology optimization method based on the vertex method, which changes the shape of the material by changing the vertex parameters of the material in each sub-region, thereby improving the performance of the motor. It solves the problems of high requirements for practitioners and dependence on the original topology in the process of existing motor design and optimization.

Technical solution: The present invention adopts a motor topology optimization method based on the vertex method in order to achieve the above-mentioned purpose of the invention, specifically as follows:

This method regards the design area of the motor as a combination of different shapes of materials, uses the coordinate parameters of each vertex of the material to represent the shape of each material, and then optimizes the coordinate parameters of each vertex, that is, the position of the vertex, to optimize the topological structure of the motor and achieve The goal of optimizing motor performance.

The motor topology optimization method specifically includes the following steps:

，

，将每个顶点的横纵坐标作为参数，共有q个参数，

；S1. In the two-dimensional coordinate system, the design area of the motor is divided into m block sub-areas, wherein the number of vertices of the i-th block sub-area is set to

,

, taking the horizontal and vertical coordinates of each vertex as parameters, there are q parameters in total,

;

S2. Connect the vertices on the edge of a certain material in the same sub-region to form a polygon, and use the polygon as the shape of a material in the sub-region, and the remaining part in the sub-region is another material; according to each The distribution of material within the subregion determines the motor topology;

S3. Select the target to be optimized according to the design requirements, and use the intelligent optimization algorithm to optimize the vertex position coordinate parameters, thereby changing the topology of the motor and obtaining the optimized new vertex position coordinates;

S4. Obtain a certain material shape in the same sub-region according to the new optimized vertex position coordinates, determine the motor topology, use the finite element method to calculate the performance of the motor with this topology, and judge whether the performance meets the design requirements. If so, then The design is completed; otherwise, return to step S1.

In the two-dimensional coordinate system, the design area of the motor is divided into m block sub-areas, and the division method is divided into uniform division and uneven division. If the design area is the stator tooth area, the uniform division method is adopted; if the design area is The rotor core is divided unevenly.

个点为顶点的多边形区域来表示其材料部分，若该材料占满该分区域，则该分区域仅有一种材料。In the m sub-regions of the motor, the i-th sub-region is used for

A polygonal area whose vertices are vertices is used to represent its material part. If the material occupies the sub-area, there is only one material in the sub-area.

Said material is silicon steel sheet, permanent magnet or air.

The optimization of the vertex position coordinate parameters using an intelligent optimization algorithm is specifically: initially generating multiple groups of random coordinate parameters x, where x is a q-dimensional vector, calculating the objective function f(x), and then according to the value of f(x) and the specific The intelligent algorithm used changes the value of x, recalculates the value of f(x), and enters the next iteration; only need to define the value range of x and the objective function f(x), and f(x) can be found automatically The minimum value of corresponds to the parameter x.

The selection of the target to be optimized according to the design requirements is to select one or more optimization targets according to the actual needs of the project, and the topology optimization problem is modeled as a single-objective or multi-objective optimization problem, which is represented by the following mathematical expression:

为每一个目标函数，n为目标函数的个数，

为限制参数x的不等式约束条件，d为不等式约束的个数，

为限制参数x的等式约束条件，k为等式约束的个数。Where x is a q-dimensional vector, y is a set of objective functions to be minimized,

For each objective function, n is the number of objective functions,

is the inequality constraint condition that limits the parameter x, d is the number of inequality constraints,

is the equality constraint condition that limits the parameter x, and k is the number of equality constraints.

为其本身；对于平均转矩、效率需要最大化的目标，

为其倒数；对于转矩、效率这些和电机电磁性能相关的目标，使用有限元法计算电机的性能；对于成本只和电机的材料使用情况有关的目标，直接根据电机拓扑确定材料的使用情况，直接计算得到。The target that needs to be optimized includes average torque, torque ripple, efficiency and cost; For the target that torque ripple and cost need to be minimized,

for itself; for the goal of maximizing average torque and efficiency,

Its reciprocal; for the goals related to the electromagnetic performance of the motor, such as torque and efficiency, use the finite element method to calculate the performance of the motor; for the goals whose cost is only related to the material usage of the motor, directly determine the usage of materials according to the topology of the motor, directly calculated.

进行优化，根据优化后的参数确定电机最终拓扑结构。The plurality of optimization objectives, obtain the motor topology according to the coordinate parameters, calculate the objective function, use the finite element method to calculate the electromagnetic performance of the motor, directly calculate the cost, and use the intelligent optimization algorithm to optimize the coordinate parameters

Carry out optimization, and determine the final topology of the motor according to the optimized parameters.

The intelligent algorithm uses simulated annealing algorithm, genetic algorithm or particle swarm algorithm.

Beneficial effect: After using the vertices to model the material shape of each sub-area, the shape of the design area will evolve with the change of parameters. During the optimization process, the shape of the design area will continuously generate different new schemes. Therefore, compared with the traditional motor design method, the present invention reduces the designer's intervention in the motor topology design, so it can achieve the purpose of improving the performance of the motor while reducing human intervention and automatically evolving the motor topology. Has the following beneficial effects:

(1) The motor designed based on this method can reduce human intervention and does not depend on the existing topology.

(2) The method is simple and feasible, and is applicable to various types of motors.

(3) According to different target performance, there are many choices, and the appropriate topology can be selected according to the needs.

Description of drawings

FIG. 1 is a flowchart of a motor topology optimization design method based on the vertex method in the present invention.

Figure 2(a) is a schematic diagram when the design area is the stator core teeth, and the division method is uniform division. Figure 2(b) is a schematic diagram when the design area is the motor rotor part, and the division mode is uneven division.

Fig. 3 is a schematic diagram of selecting the rotor as the design area, dividing the motor rotor area and selecting the number of vertices.

Fig. 4 is a schematic diagram of the rotor at the beginning of optimization, during optimization and after optimization.

Figure 5(a) is a schematic diagram of the average torque and fluctuation of the entire genetic algorithm optimization process as the optimization algebra changes, and Figure 5(b) shows the distribution of all optimization results and the Pareto front.

Figure 6(a) is the topology corresponding to point A in Figure 5(b), Figure 6(b) is the topology corresponding to point B in Figure 5(b), and Figure 6(c) is the topology in Figure 5(b) The topology corresponding to point C, Figure 6(d) is the topology corresponding to point D in Figure 5(b).

Detailed ways

The technical scheme of the invention will be described in detail below in conjunction with examples. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings but not all of them.

As shown in FIG. 1 , it exemplarily shows a flow chart of the motor topology optimization method based on the vertex method disclosed in the present invention.

，

，将每个顶点的横纵坐标作为参数，共有q个参数，

；S1. In the two-dimensional coordinate system, the design area of the motor is divided into m block sub-areas, wherein the number of vertices of the i-th block sub-area is set to

,

, taking the horizontal and vertical coordinates of each vertex as parameters, there are q parameters in total,

;

S2. Connect the vertices on the edge of a certain material in the same sub-region to form a polygon, and use the polygon as the shape of a material in the sub-region, and the remaining part in the sub-region is another material; according to each The distribution of material within the subregion determines the motor topology;

S3. Select the target to be optimized according to the design requirements, and use the intelligent optimization algorithm to optimize the vertex position coordinate parameters, thereby changing the topology of the motor and obtaining the optimized new vertex position coordinates;

S4. Obtain a certain material shape in the same sub-region according to the new optimized vertex position coordinates, determine the motor topology, use the finite element method to calculate the performance of the motor with this topology, and judge whether the performance meets the design requirements. If so, then The design is completed; otherwise, return to step S1.

The finite element method comes with the maxwell software. After drawing the motor model, you can directly use the finite element method to calculate.

An example of optimizing a synchronous reluctance motor is as follows:

1. The synchronous reluctance motor rotor is only composed of silicon steel sheets and air. In this example, the materials are only silicon steel sheets and air. Considering the symmetry, only the 1/8 model needs to be considered. As shown in Figure 3, the design area of the rotor of the motor is divided into 8 sub-areas. Considering the symmetry, only the left 4 parts are considered. The air shape of each part is represented by 6 vertices, and each vertex has 2 parameters, so there are 48 optimized parameters in this example.

2. Connect the vertices in the same sub-area to form a polygon, and use the polygon as the shape of the air in the sub-area, leaving some silicon steel sheets in the sub-area; determine the motor topology according to the distribution of materials in each sub-area.

3. In this example, the optimization target is selected as the average torque and torque fluctuation, and the genetic algorithm is used for optimization. In the optimization process, a new generation of population is first generated by genetic algorithm (the first generation of population is the coordinate parameters of multiple groups of vertices); then the corresponding topological motor is drawn in the maxwell software, and the performance of the motor is calculated; the high-quality population is retained according to the performance, and Perform crossover on high-quality populations (replace and recombine part of the vertex parameters of two high-quality individuals in the population to generate new individual parameters), mutation operations (change the values of some coordinate parameters of individual strings in the population), and enter the next Iterate until the maximum algebra limit is reached.

4. After the optimization is completed, the motor topology is determined according to the new optimized coordinate parameters. Verify that other indicators meet the design requirements. If satisfied, the optimization is completed; otherwise, return to the first step.

Fig. 4 exemplarily shows the evolution process of the motor rotor during the optimization process.

Figure 5(a) exemplarily shows that when the genetic algorithm is used as the optimization method in the motor topology optimization method based on the vertex method, and the objective function is selected as the average torque and torque ripple, the average torque and torque fluctuation in each iteration process A plot of the mean value of moment fluctuations as the number of iterations increases. Fig. 5(b) exemplarily shows the distribution and Pareto front of all motor performances in a complete optimization process. It can be seen from the figure that the average torque of the motor can be increased and the torque fluctuation can be reduced by adopting the present invention.

Figure 6(a), Figure 6(b), Figure 6(c), and Figure 6(d) exemplarily show the motor rotor topology corresponding to the points on the Pareto front.

The motor topology optimization method based on the vertex method disclosed in the present invention can obtain different results according to the selection of the objective function, and further determine the final topology according to the actual situation of the motor, other performance requirements and process technology.

The above embodiments are only exemplary descriptions of this patent, and do not limit its protection scope. Those skilled in the art can also make partial changes to it, as long as they do not exceed the spirit and essence of this patent, they are all within the protection scope of this patent.

Claims (10)

1. A motor topology optimization method based on the vertex method, characterized in that, the method regards the design area of the motor as a combination of different shape materials, represents the shape of each material with each vertex coordinate parameter of the material, and then passes the optimization method Optimize the coordinate parameters of each vertex, that is, the vertex position, so as to optimize the topology of the motor and achieve the goal of optimizing the performance of the motor. 2. the motor topology optimization method based on vertex method according to claim 1, is characterized in that, described motor topology optimization method specifically comprises the steps: S1. In the two-dimensional coordinate system, the design area of the motor is divided into m block sub-areas, wherein the number of vertices of the i-th block sub-area is set to

,

, taking the horizontal and vertical coordinates of each vertex as parameters, there are q parameters in total,

, the value range of the parameter cannot exceed the range of the design area; S2. Connect the vertices on the edge of a certain material in the same sub-region to form a polygon, and use the polygon as the shape of a material in the sub-region, and the remaining part in the sub-region is another material; according to each The distribution of material within the subregion determines the motor topology; S3. Select the target to be optimized according to the design requirements, and use the intelligent optimization algorithm to optimize the vertex position coordinate parameters, thereby changing the topology of the motor and obtaining the optimized new vertex position coordinates; S4. Obtain a certain material shape in the same sub-region according to the new optimized vertex position coordinates, determine the motor topology, use the finite element method to calculate the performance of the motor with this topology, and judge whether the performance meets the design requirements. If so, then The design is completed; otherwise, return to step S1. 3. the motor topology optimization method based on vertex method according to claim 2, is characterized in that, in the two-dimensional coordinate system, the design area of the motor is divided into m block sub-regions, and the division mode is divided into uniform division and non-uniform division. Evenly divided, if the design area is the stator tooth area, the uniform division method is adopted; if the design area is the rotor core, the uneven division method is adopted. 4. the motor topology optimization method based on vertex method according to claim 3, is characterized in that, in the m block sub-areas of the motor, the i-th block sub-area is used for

A polygonal area whose vertices are vertices is used to represent its material part. If the material occupies the sub-area, there is only one material in the sub-area. 5. The motor topology optimization method based on the vertex method according to claim 4, wherein the material is silicon steel sheet, permanent magnet or air. 6. The motor topology optimization method based on the apex method according to claim 2, wherein said use of an intelligent optimization algorithm to optimize the apex position coordinate parameters is specifically: initial generation of multiple groups of random coordinate parameters x, where x is q-dimensional vector, calculate the objective function f(x), then change the value of x according to the value of f(x) and the specific intelligent algorithm used, recalculate the value of f(x), and enter the next iteration; only need to define The value range of x and the objective function f(x) can automatically find the parameter x corresponding to the minimum value of f(x). 7. The motor topology optimization method based on the vertex method according to claim 2, characterized in that, the selection of the target to be optimized according to the design requirements is to select one or more optimization targets according to specific engineering actual needs, and the topology The optimization problem is modeled as a single-objective or multi-objective optimization problem, represented by the following mathematical expression:

Where x is a q-dimensional vector, y is a set of objective functions to be minimized,

For each objective function, n is the number of objective functions,

is the inequality constraint condition that limits the parameter x, d is the number of inequality constraints,

is the equality constraint condition that limits the parameter x, and k is the number of equality constraints. 8. according to claim 2 or 7 described motor topology optimization methods based on vertex method, it is characterized in that, the target that needs optimization comprises average torque, torque fluctuation, efficiency and cost; For torque fluctuation, cost needs the goal of minimizing,

for itself; for the goal of maximizing average torque and efficiency,

Its reciprocal; for the goals related to the electromagnetic performance of the motor, such as torque and efficiency, use the finite element method to calculate the performance of the motor; for the goals whose cost is only related to the material usage of the motor, directly determine the usage of materials according to the topology of the motor, directly calculated. 9. The motor topology optimization method based on the vertex method according to claim 7, characterized in that, the multiple optimization targets obtain the motor topology according to the coordinate parameters, calculate the objective function, and use the finite element method to calculate the electromagnetic performance of the motor, directly Calculate the cost and use the intelligent optimization algorithm to adjust the coordinate parameters

Carry out optimization, and determine the final topology of the motor according to the optimized parameters. 10. The motor topology optimization method based on the vertex method according to claim 8, wherein the intelligent algorithm uses a simulated annealing algorithm, a genetic algorithm or a particle swarm algorithm.

Images