Hierarchical predictive energy management strategy for fuel cell buses entering bus stops scenario

Abstract

This paper aims to answer how to use traffic information to design energy management strategies for fuel cell buses in a networked environment. For the buses entering the bus stops scenario, this paper proposes a hierarchical energy management strategy for fuel cell buses, which considers the traffic information near the bus stops. In the upper-level trajectory planning stage, the optimal SOC trajectory under various historical traffic conditions is solved through dynamic planning. The traffic information and the best SOC trajectory are mapped through BiLSTM, which can achieve fast, real-time long-term SOC reference. In the lower-level real-time predictive energy management strategy, the optimal SOC is used as the state reference to guide the predictive energy management of fuel cell buses when entering the bus stops. Simulation results show that compared with the strategy without SOC trajectory reference, the life cost of the proposed strategy is reduced by 13.8%, and the total cost is reduced by 3.61%. The SOC of the proposed strategy is closer to the DP optimal solution.