Stochastic optimal control for hybrid electric vehicles running on fixed routes

Fixed-route driving is very common in real world, and it is different from the fixed-cycle driving in which no uncertainties are included. However, most hybrid electric vehicle (HEV) energy management strategies are developed under fixed cycles and there is not guarantee of performance of these strategies under the real-world driving conditions. The knowledge of fixed-cycle HEV optimal control usually cannot be directly extended to fixed-route driving. In this paper, a stochastic optimal control approach for fixed-route HEV is presented. The historical data on the fixed route are utilized and a road-segment-based discrete stochastic model is constructed. The energy management optimization problem is solved using stochastic dynamic programming. Most of the computation tasks can be conducted off-line, so this method can be used for onboard implementation in real-time. A time-varying scaling method is used to generate fixed-route driving data in simulation based on a standard driving cycle. The simulation results show that the performance of the proposed stochastic optimal control strategy consumes only 1.0% more energy than the global optimal result after 24 trips on the fixed-route and outperforms the other real-time HEV energy management strategies.