An optimal structure selection and parameter design approach for a dual-motor-driven system used in an electric bus

•A systematic bi-level design framework for a dual-motor driven system was developed.•The scalable models including gear system, wet clutch and motors were built.•Dual-motor driven system with two clutches shows better efficiency than one clutch.•Decreasing the power level can improve the efficiency of dual-motor driven system.

A number of driving system topologies have been developed for electric vehicles, but the topology design and optimal sizing always challenge the performance of electric vehicles. This paper attempts to address three aspects. First, two novel topologies were derived from the original dual-motor-driven system, and the efficiency models of the system components were built, including motor efficiency, planet gear system efficiency and drag loss of the wet clutch. Second, a systematic optimal sizing framework was constructed. The feasible region of the design parameters was divided into a certain number of grid points, and each grid point represented different design results. Then, a dynamic programming algorithm was applied to each grid point to locate the optimal control strategy and obtain the best grid points under different power levels. After that, a bi-level optimization method was applied at these selected grid points to find the optimal design parameters. Last, the simulation results showed that, compared with the original design, the new topology with two clutches can reduce energy loss by 12.4%, and the optimal design results for the original topology can reduce energy loss by 3.36%.