A model predictive control (MPC) based robust energy management strategy for power-split hybrid electric vehicles (HEVs) is proposed in this paper which can guarantee charge sustaining constraint satisfaction in presence of uncertainty in future torque demand and vehicle velocity. The proposed method utilizes robust backward reachability analysis to pre-compute a robust set from where control solutions exist that ensure charge sustaining constraint satisfaction despite the uncertainty. This set is then used as the terminal set constraint of the MPC problem followed by tightening of state constraints for robust constraint satisfaction. An equivalent convex optimal control is then formulated which is solved in a receding horizon fashion. Simulation results show the efficacy of the proposed control strategy in robust constraint satisfaction.
Stochastic model predictive control for energy management of power-split plug-in hybrid electric vehicles based on reinforcement learning
