Method and device for optimizing structural parameters of switched reluctance motor

Abstract

The invention provides a switch reluctance motor structure parameter optimization method and device and belongs to the technical field of switch reluctance motors. The method comprises the steps of calculating initial structure parameters of a switch reluctance motor (SRM); selecting to-be-optimized structure parameters from the initial structure parameters of the motor obtained in the first step; establishing a motor finite element model for the initial structure parameters obtained in the first step and acquiring the performance parameters, the efficiency eta and the torque pulsation coefficient delta, of the motor through simulation; establishing sample data according to the to-be-optimized structure parameters and the performance parameters obtained in the second step and the third step; according to the sample data in the fourth step, training the sample data by employing the fruit fly optimization algorithm-extreme learning machine (FOA-ELM) algorithm to obtain a to-be-optimized switch reluctance motor model; according to the switch reluctance motor model obtained in the fifth step, with the to-be-optimized structure parameters of the motor as the optimization objects and the efficiency eta and the torque pulsation coefficient delta as the optimization target, performing optimization by using the diminishing step chaotic mapping fruit fly algorithm to obtain the optimal structure parameters of the switch reluctance motor. The method and the device have the advantages of high motor structure parameter optimization speed and optimization precision.

Description

technical field [0001] The invention relates to the technical field of switched reluctance motors, in particular to a method and device for optimizing structural parameters of switched reluctance motors. Background technique [0002] Switched reluctance motor has been widely used in many fields because of its advantages of small starting current, large starting torque, simple structure and low cost. However, due to its double salient pole structure, nonlinearity and saturation effect of the magnetic circuit, it has the problem of large torque ripple, which affects its popularization and application. For this reason, researches have been carried out on optimizing the structural parameters of switched reluctance motors at home and abroad, and various optimization algorithms such as simulated annealing algorithm, artificial neural network, and genetic algorithm have been proposed. Although some effects have been achieved, there are still deficiencies. For example, simulated an...

AI Technical Summary

Problems solved by technology

Compared with the traditional optimization method, the weighted and multi-objective optimization algorithm has advantages in optimizing multiple objectives at the same time, but it is easy to fall into a local optimal solution; the non-elite multi-objective genetic algorithm is not stable; the cultural particle swarm algorithm The particle swarm collaborative optimization algorithm has many adjustment parameters, the workload is heavy, and the optimization efficiency is low

The above multi-objective optimization method solves the multi-objective, multi-variable, and multi-constraint problems of the switched reluctance motor to a certain extent, but it is easy to fall into the local optimal solution during the optimization process, and has the disadvantages of low efficiency and low stability. Therefore, It is of great significance to study an algorithm that can quickly converge and achieve multi-objective global optimization for switched reluctance motors

[0003] The name of the Chinese patent document is: A method for optimizing the structural parameters of a single-winding magnetic levitation switched reluctance motor, and the invention patent application number: 201410499891. The non-dominated sorting genetic algorithm optimizes the suspension force and torque at the same time, but there are many algorithm adjustment parameters and complex calculations.

Images