Design of optimal, robust energy management strategy for a parallel HEV

A novel approach based on stochastic dynamic programming (SDP) is proposed to develop optimal, robust energy control strategies for a parallel hybrid electric vehicle (HEV). Unlike other approaches that take a specific drive cycle as prior deterministic information, we model the transport mission as a stochastic process based on collected data. The problem of dynamic energy management in HEV system is formulated as an infinite-horizon SDP optimization problem and solved using an efficient "policy iteration" algorithm. Our approach guarantees an optimization of vehicle performance and is robust to the accuracy of available information. In addition, the resulting energy management strategy is suitable for real-time implementation. Hence, it is more efficient and more useful in practice. Experimental results demonstrate the effectiveness of our approach compared to a rule-based algorithms.