Energy management of hybrid electric powertrain using predictive trajectory planning based on direct optimal control

Optimal control methods are being employed to solve the energy management problem between the energy convertors of hybrid automotive power trains. The focus thereby was on application or extensions of Pontryagin minimum principle and/or Hamilton-Jacobi-Bellman theory This paper approaches it as a trajectory optimization problem relying on predictive information within a limited time horizon and solves it by applying direct optimal control in a simultaneous discretization framework.In case of a parallel hybrid configuration, the energy consumption minimization problem reduces to determining engine on/off state and torque split between the combustion engine and the electric machine. A primal active set method together with heuristics on value function iterates is implemented to solve the mixed boolean convex problem. The effectiveness of the approach is demonstrated in case of a plug-in hybrid electric vehicle with a 2 cylinder combustion engine and electric machine in parallel configuration. The presented approach is investigated for optimality using an implementations of deterministic dynamic programming (DP) and analysed for realtime capability. The results from the simulation study show state trajectories close to global numerical optimum from DP with computational complexity which can be handled in real-time.