Multi-Objective Optimization Design of Rotor Parameters of External Rotor Synchronous Reluctance Machine Parameters Based on Mixed Surrogate Model

Aiming at improving the torque performance of the external rotor synchronous reluctance machine, this paper establishes a mixed surrogate model on the basis of analyzing various design parameters and performing multi-objective optimization. Firstly, the influence of parameters on the average electromagnetic torque and torque ripple ratio of the machine is analyzed. In order to improve the prediction accuracy and stability, shorten the optimization cycle, a novel mixed surrogate model based on the Kriging model and RBF model is proposed, the global accuracy and local accuracy of the model are verified, the addition criterion based on the dynamic dimension search algorithm is designed to improve the accuracy of the model, a dual optimization framework of constructing the surrogate model-optimizing model accuracy-optimizing torque characteristics is built, finally the electromagnetic torque and torque ripple ratio of the machine are optimized. The results show that the average torque and torque ripple ratio of the optimized external rotor synchronous reluctance machine are significantly improved compared with the initial design.