Multiobjective Optimization Method to Maximize Power Density and Efficiency of an Electric Propulsion Motor in UAV Applications
Jinquan Xu, Wenbo Jin, Huapeng Lin, Boyi Zhang, Hong Guo
Abstract
This paper proposes a new multi-objective optimization method to maximize the power density and efficiency of an electric propulsion motor in UAV applications. The high power density permanent magnet electric propulsion motor (PM-EPM) topology is firstly proposed with the multilayer wavy heat dissipation structure to improve the motor cooling performance. Then the electromagnetic-thermal analytic model of the PM-EPM is proposed, which can accurately predict the electromagnetic and thermal performance under various conditions. To enhance the global optimization performance, a novel multi-objective dynamic neighbourhood genetic learning strategy based on particle swarm optimization (MODNGL-PSO) method is proposed to optimize the power density and efficiency of the PM-EPM, which has low computation burden, high optimization efficiency and global optimization performance. Finally, a 20kW 3000rpm PM-EPM is designed and manufactured. The experimental results show that the proposed motor has the excellent electromagnetic performance with the rated power density of 2.25kW/kg and efficiency of 95.52%.