Research on energy-saving adaptive optimization of hybrid electric vehicle based on improved dynamic programming and control rule extraction

Developing reasonable energy management strategy(EMS) to coordinate the working status of various power sources is the key to ascendancy the energy-saving advantages of power-split hybrid electric bus. However, few optimal control strategies have been conducted to utilize operating condition information to achieve approximate online global optimization of energy-saving algorithms, and maximize energy-saving potential. Aiming at the problem, a real-time adaptive EMS based on improved DP algorithm and control rule extraction is proposed for a power-split HEB in this paper. An improved DP algorithm based on secant method for solve the lack of theoretical basis for determining weight coefficients is constructed. The optimal control rules are extracted through outlier detection and graphical method. And an online adaptive algorithm within Relevance Vector Machine and particle swarm optimization algorithm is performed to achieve online adaptive optimal effect. Finally, the simulation and hardware-in-the-loop test are conducted. Simulation results validate that control algorithm proposed in this paper is close to DP results, while achieving the 14.6% oil saving compared with logic threshold control. Additionally, the control algorithm proposed in this paper also demonstrates good real-time performance. The main contribution of this paper is to explore a novel way to fully exert the fuel economy potential and adaptability of power-split HEB in various cycles.