Predictive energy management of series-parallel hybrid electric vehicles and its sensitivity analysis to dynamic traffic environment

This paper proposes a efficient model predictive controller (MPC) for the energy management problem of dual mode intelligent (DM-i) hybrid powertrain equipped vehicles in navigation driving scenarios. The controller is divided into two layers where the upper-layer plans the optimal state of charge (SOC) trajectory according to the dynamic traffic information provided by the navigation system, while the lower-layer determines the optimal operation mode and engine working points. The alternating direction multiplier method (ADMM) is employed in the upper-layer to obtain the optimal solution considering mode switching. Simulation results show that compared with the rule-based strategy, MPC operates the powertrain with 1.48% and 2.49% lower fuel consumption in urban and highway routes, respectively. In the complex routes composed of urban and highway, the fuel-saving rate of MPC is 6.2% and 8.92%, respectively. In the varying scenarios where the driver temporarily changes the driving route, the economy of MPC is still better than the rule-based strategy, which ensures that MPC has strong robustness in practical applications.