Energy management strategy is vital for a plug-in hybrid electric vehicle and in this paper, a strategy based on stochastic model predictive control is proposed. Firstly, Markov Chain Monte Carlo Simulation is used to predict velocity sequences in the 10-second horizon followed by post-processing like average filtering, quadratic fitting, etc. which is meant to moderate fluctuations of the results. The RMSE is controlled around 2.4357 Km/h. Moreover, dynamic programming is adopted to construct a benchmark strategy and also to act as the rolling optimization part of SMPC-based strategy. The results show that the fuel economy of the strategy based on SMPC is around 13 percent worse than that on DP. However, with 14.7 L/100 km as fuel consumption, it is still within reasonable ranges.
Xie, S., Peng, J., & He, H. (2017). Plug-In Hybrid Electric Bus Energy Management Based on Stochastic Model Predictive Control. In Energy Procedia (Vol. 105, pp. 2672–2677). Elsevier Ltd. https://doi.org/10.1016/j.egypro.2017.03.773