Cuckoo Search Algorithm for Multi-Objective Optimization of Transient Starting Characteristics of a Self-Starting HVPMSM

Abstract

The starting and synchronization process of high-voltage permanent magnetic synchronous motors (HVPMSM) is a complex electromechanical transient problem. This paper proposes a multi-objective five-dimensional optimization method for starting characteristics of HVPMSM based on quadratic response surface models (QRSM) combined with Cuckoo search algorithm. Eighty-nine sets of transient finite element models were developed. To obtain starting current, starting time, starting torque and torque ripple, the transient process of HVPMSM is calculated with different cage bar width, cage bar depth, cage bar conductivity, and slot wedge conductivity. Steady-state and locked rotor experiments are conducted to verify the accuracy of the FEA. Then, the Cuckoo search method is employed to optimize design parameters achieved by the finite element models to obtain the minimum starting current, the maximum starting torque, the shortest starting time, and the minimum torque ripple, while maintaining high output performance and low cost. The results demonstrate that QRSM combined with Cuckoo search algorithm accurately describe the non-linear relationship between the structural characteristics, material properties and output performance of the HVPMSM. The best solutions are consistent with the results of the finite element analysis. In addition, the theoretical method is verified by experiments.