Parameter closed-loop optimization for pure electric vehicles: unified design of power system and control parameters

摘要

To solve the problem of unreasonable vehicle parameters caused by unknown curb mass and open-loop power system and control strategy optimization in the development of pure electric vehicles, this paper presents a parameter closed-loop optimization algorithm (COA) via unified design of system and control parameters. First, a mass closed-loop algorithm (MCA) was adopted to optimize the parameters of power systems under an unknown curb mass and its convergence was proven. Next, with energy consumption being the index function, the torque distributions of the front and rear motors were optimized by dynamic programming (DP). Additionally, vehicle power system and control strategy parameter optimization were realized by combining the MCA, DP, and genetic algorithm. Finally, two comparative optimization algorithms which are the assumed curb mass optimization algorithm (AOA) and the torque equal ratio distribution optimization algorithm (TOA) were implemented to validate the proposed algorithm. The simulation results under China light-duty vehicle test cycle-passenger car (CLTC-P) and New European Driving Cycle (NEDC) operating conditions indicate that the proposed algorithm achieves minimum energy consumption compared with the competing optimization algorithms. The energy consumption resulting from the COA is reduced by 18.98 and 6.36, compared with those of the TOA and AOA, respectively, under CLTC-P conditions. The energy consumption of the COA is 16.57 and 6.91 lower than those of the TOA and AOA, respectively, under NEDC conditions. This algorithm can optimize the curb mass, peak powers of the motors, mass of the motors, battery energy, battery mass, and torque distribution coefficient simultaneously, and can be applied to different operating conditions.