Optimal Design Of Surface Permanent Magnet Motor Based On A Improved Kriging Model

With the development of economy and the progress of science and technology,the performance parameters accuracy of electromagnetic equipment is required higher and higher in practical engineering applications.The relationship between the performance parameters and the influence variables of electromagnetic equipment is usually complex and nonlinear,and there are many local optimal points.At present,the direct combination of traditional optimization algorithm and finite element analysis requires a large number of electromagnetic field numerical simulation,which is expensive,time-consuming and prone to local optimality.Nowadays,more and more electrical scholars pay attention to the combination of surrogate model and traditional optimization algorithm because it can reduce the cost of computation and improve the speed and precision of search.Especially in the complex multi-objective practical engineering optimization problem,it is of great significance to choose appropriate surrogate model and efficient optimization algorithm to reduce the amount of calculation and improve the accuracy.To solve these problems,a new global optimization algorithm based on improved Kriging surrogate model is proposed in this thesis.After analyzing the performance of the algorithm,it is applied to multi-objective optimization.Firstly,an improved Kriging model based global optimization algorithm with high efficiency and precision was proposed.Based on the comparison of the advantages and disadvantages of the existing optimization algorithms,the Kriging surrogate model was studied deeply,and the classification of the Kriging surrogate model was given according to the basis function.Combining the improved Kriging surrogate model with multi-objective particle swarm optimization,the improved Kriging model was used to approximate the objective function and constraint function in PSO.And the corresponding program was compiled on MATLAB.Secondly,the performance of optimization algorithm based on improved Kriging model was verified.In order to explore the superiority of the optimization algorithm based on the improved Kriging model,this thesis selected five standard test functions to test the performance of the algorithm.First,two simple test functions were used to verify the RMSE of the two common Kriging models and the Kriging model used in this thesis,and then two standard test functions with different characteristics were used to verify the optimization ability of the proposed algorithm and the general particle swarm optimization algorithm.Then,two multi-objective standard test functions were used to verify the accuracy and optimization performance of the optimization algorithm based on the improved Kriging model.Finally,the proposed global optimization algorithm was applied to the multi-objective optimization design of a surface permanent magnet synchronous motor.Based on the FEM simulation experiment,the parameters that affect the cogging torque and the amount of magnets were analyzed,and the improved optimization algorithm was used to reduce the cogging torque.