A Stochastic Model Predictive Control Strategy for Extended Range Electric Vehicle

Abstract

The Energy Management Strategy(EMS) of the Extended-Range Electric Bus (E_REB) plays an important role in the fuel economy and emission performance. This paper presents a Stochastic Model Predictive Control (SMPC) strategy based on Stochastic Dynamic Programming (SDP) and Model Predictive Control (MPC) for E_REB. SMPC applies Markov stochastic prediction model to predict the power demand of driving cycles, converts the SDP algorithm into the local optimization algorithm within the finite horizon, then the global suboptimal strategy can be obtained by solving a finite horizon optimization problem. SMPC can greatly save the online computation and operation time, achieve the goal of real-time online control. Simulation results of E_EVB are presented to demonstrate the effectiveness of the SMPC.