Design of an Energy Management Strategy for a Parallel Hybrid Electric Bus Based on an IDP-ANFIS Scheme

A critical challenge in the energy management of hybrid electric vehicles is how to intelligently distribute the power between engine and electric motor with a significant reduction in fuel consumption and emissions. In this paper, iterative dynamic programming (IDP) and adaptive neurofuzzy inference system (ANFIS)-based energy management strategy was proposed for a parallel hybrid electric bus. First, IDP was used to obtain the optimal control trajectories for a specific driving cycle. During the iterative process, a modified gear-shifting strategy was introduced to achieve a good tradeoff between the fuel economy and the drivability performance. Next, multi-ANFIS networks were designed and trained to learn the control law from the available optimal trajectories. Finally, the real-time energy management controller based on the IDP-ANFIS was built to coordinate the output of the two power sources, which reduces the fuel consumption and emissions. The simulation and experimental results reveal that the IDP-ANFIS is feasible, and the performance is superior to that of the equivalent consumption minimization strategy and rule-based method.