An Optimized Design Method of Homopolar Inductor Alternator Based on Genetic Algorithm

Abstract

Homopolar inductor alternator (HIA) plays an important role in capacitor charging power supply (CCPS) system. Due to the 3-D-structure magnetic circuit and short-circuit working condition, the design method of HIA is different from that of conventional machines obviously. On the current research, the traditional design method is usually adopted, that is the alternator is designed depending on experience at first, and then, the charging performance is verified by system simulations. This method could cause HIA’s performance to be much higher than the system’s requirement, resulting in energy wasting, in addition, too big system is not suitable for some size-limited situations, such as on-board, ship-borne, and other mobile supply. In this article, a new optimization design method of HIA is proposed. An “optimal function” is derived first, which can find out the “optimal charging power point” under the charging requirements. Then, genetic algorithm (GA) is applied to investigate the HIA with smaller volume and lighter mass. Finally, in order to verify the optimization method, the characteristics of the optimal HIA are carried out by finite element method and system simulation. The results demonstrate that the designed HIA can well meet the charging requirements, and its performance is better than the HIA designed by the traditional method obviously. Meanwhile, the proposed design method has significant advantage to design smaller and lighter alternators, which is important to improve the output performance and mobility of HIA-CCPS.