Research On Sequential Multi-Objective Complete Factor Analysis Optimization Method Of Permanent Magnet Synchronous Servo Motor

Servo permanent magnet synchronous servo motor is an important actuator of vehicle unmanned control system.It has the advantages of reliable operation,high working efficiency,strong carrying capacity and easy control.The multi-objective optimization design method of permanent magnet synchronous motor based on finite element calculation is an important technical means to improve the working performance of the motor body.However,the traditional permanent magnet synchronous motor multiobjective optimization method based on the principle of control variables has the disadvantages of long simulation period,low optimization efficiency,and difficulty in achieving optimal design.Aiming at the problems of the above traditional motor optimization methods,this paper proposes a sequential multi-objective full factorial optimization method for permanent magnet synchronous servo motors based on factor sensitivity analysis to improve the multi-objective optimization efficiency and target expected value of the motor and shorten the test cycle.The main research contents of this article are as follows:Servo permanent magnet synchronous motors usually use fractional slot concentrated winding structure to improve the performance of the servo system.Due to the high harmonic components of the stator and rotor magnetic fields of the motor,it is easy to produce low-order radial electromagnetic force waves,causing certain vibration and noise problems.Summarize the force wave order,source and action form of the radial electromagnetic force wave of the induction synchronous motor through theoretical derivation.Based on the method of harmonic magnetic field analysis,the radial force wave distribution between the stator side,the rotor side and the fixed rotor is calculated.From the aspects of winding coefficient,cogging torque and radial electromagnetic force wave order,the appropriate motor pole slot matching scheme is selected to theoretically suppress the electromagnetic noise and torque ripple of the motor.Aiming at the problems of high degree of coupling of the design factors of permanent magnet synchronous servo motors and the difficulty of visualizing the target response,the parametric modeling of permanent magnet synchronous motors under multi-variable interactive response is implemented using response surface method,which provides a model reference for the multi-objective optimization process of the motor.According to the result of factor sensitivity analysis,the response surface nonlinear function model of multi-objective high sensitivity factor is calculated by the design method of central surface sequential experiment.For the servo permanent magnet synchronous servo motor optimized in this paper,in order to improve the data fitting accuracy of the traditional response surface modeling method,this paper proposes a mixed response surface modeling method based on improved test space.By adding suitable space test points,the purpose of improving the fitting accuracy of the model is achieved.According to the results of factor sensitivity analysis,the response surface nonlinear function model of multi-objective high sensitivity factor is calculated by the design method of central surface sequential test.In order to improve the data fitting accuracy of traditional response surface modeling methods,a mixed response surface modeling method based on function model and test space improvement is proposed.By adding appropriate space test points,the purpose of improving design space utilization rate and model fitting accuracy is achieved.Because the traditional multi-objective particle swarm optimization algorithm has problems such as weak global optimization ability and easy to fall into regional advantage solutions,a mixed particle swarm multi-objective optimization algorithm based on genetic mutation is proposed.Using linearly decreasing mutation rate and the selection principle of the global optimal solution "minimization of position variance" to reduce the iterative calculation amount of the algorithm,improve its convergence speed and global retrieval ability;the principle of genetic mutation is used to realize the updating iteration of the population,and the non-inferior solution set of the multi-objective optimization problem of the permanent magnet synchronous servo motor is accurately and quickly solved.Aiming at the problems of long optimization period and difficulty in achieving optimal design in traditional motor multi-objective optimization methods based on finite element calculation.This paper proposes a multi-objective optimization strategy based on factor sensitivity analysis and sequential experiment design for permanent magnet synchronous servo motors with complete factoring.According to the spatial distribution law of the sensitivity of the design factors,it is optimized in stages.By selecting appropriate optimization strategies for each level of design factors,the high-efficiency,complete factorial multi-objective optimization design of the permanent magnet synchronous servo motor is achieved.