Multi-Objective Optimization of Gear Ratios of a Seamless Three-Speed Automated Manual Transmission for Electric Vehicles Considering Shift Performance

Abstract

Multi-speed transmission can greatly improve the power and economic performance of electric vehicles (EVs) compared with single-speed transmission. Gear ratio is the key design parameter of multi-speed transmission. Optimizing gear ratios can further improve vehicle performance. Most of the existing optimization methods of gear ratios take the power and economy of vehicles in gear as the optimization objectives, but rarely consider the shift performance of the transmission, such as shift time, friction, and shift jerk. Considering the shift performance in the process of gear ratio optimization can not only optimize the vehicle performance in gear, but also improve the shift performance of the transmission. Therefore, this paper proposes a multi-objective optimization method of gear ratios considering the shift performance. Firstly, a seamless three-speed automated manual transmission (AMT) of EVs is selected as the research object, the structure and the shift process without power interruption of the three-speed AMT are introduced, and the detailed EV simulation model is established. Then, the multi-objective optimization method of gear ratios considering shifting performance is described. Specifically, the acceleration time, energy consumption, and jerk of the vehicle in gear are taken as the objective functions, and the shift time, clutch friction, and the shift jerk are added to the corresponding objective functions, respectively. Finally, the multi-objective optimization algorithm is used to solve the gear ratio optimization problem. The simulation results show optimization of the gear ratios significantly improves the power, economy, and comfort of the vehicle compared with the original. More importantly, compared with the optimization method without shift performance, gear ratios optimized by the proposed optimization method has better shift performance, and the feasibility of the proposed method is verified by simulations.