A Global Optimization Approach for Electrical Machine Designs

Optimization techniques have been widely applied to the design of electric machines. Traditional approaches, however, are only effective to find the local optima. This paper reports an engineering project that concerns the global optimization issues of electrical machine design. An advanced mathematical method, the filled function method (FFM), is adapted to seek the global minimum of the multi-modal model of electrical machines. Detailed analysis is presented for the configuration of the FFM in a constrained environment, construction of the algorithm, and implementation of the software. Results of the numerical experiments shows that the effectiveness of the proposed approach.