Optimal Design of a PMSM for Electric Vehicle Using Chaotic Particle Swarm Optimization

In this paper, an optimal design approach for a permanent magnet synchronous machine (PMSM), as an electric vehicle (EV) traction machine is proposed in such a way that a high power density machine can be achieved. A surface mounted PMSM is selected for its high efficiency and torque density. The machine parameters and the design constraints are established based on the EV characteristics. Guided by the analytical results of the design approach, a finite element study is performed on the machine to verify the relevance of the proposed approach. Then, a new optimization method, combining particle swarm optimization (PSO) and chaos theory, is applied for the machine design parameters in order to maximize the power density. This algorithm, referred to as CPSO, utilizes the chaos system properties such as its sensitivity to the initial conditions and its dynamical behaviour to enhance the global search capabilities of the PSO algorithm. The simulation results show the effectiveness of this algorithm compared to other meta-heuristic algorithms in increasing the machine power density. Furthermore, it is found that the optimized PMSM offers a 41.60% improvement in power while respecting the EV characteristics.