Stochastic Model Predictive Control Strategy with Short-term Forecast Optimal SOC for a Plug-in Hybrid Electric Vehicle

Abstract Both the stochastic traffic information and State of Charge (SOC) are great impact on the performance of the plug-in parallel hybrid electric vehicle (PHEV). The application of velocity prediction offers the forthcoming trip information directly to estimate the driving power and arrange the SOC. To address this issue, a Stochastic Model Predictive Control (SMPC) based energy management strategy ( EMS) considering short-term forecast optimal SOC is proposed. Firstly, a multiple linear regression of engine and battery is developed for SMPC oriented application, respectively. Then, the velocity prediction model is developed based Markov Chain and the reference SOC optimized by Dynamic Programming (DP) using the forthcoming information. On this basis, the SMPC based EMS with the short-term optimal SOC is constituted. Finally, the various prediction horizon and driver styles are discussed to validate the proposed strategy. The normal MPC and the DP are used as benchmark strategies for comparison to evaluate the evolution of short-term optimal SOC and performance of SMPC based EMS. The test results indicate that the SMPC with the short-term optimal SOC made it possible to promote the EMS with capable of significantly improving the fuel economy of the plug-in PHEV.