A Novel Model Predictive Control Framework for Energy Management in Retrofit Hybrid Electric Vehicles

Hybrid Electric Vehicles (HEV) are increasingly gaining focus and usage for their ability to effectively reduce fuel consumption and emissions. In retrofit HEVs, additional electrical power components are retrofitted to the existing fuel-powered engine-based conventional vehicles which provide an easier and more economical means to transform them into HEVs. In this work, a novel control strategy is developed for the energy management of a retrofit mild parallel HEV where there is neither any control over the engine system nor direct sensing of engine variables. The energy management–based control strategies of a Model Predictive Control (MPC) and Equivalent Consumption Minimization Strategy (ECMS) are analyzed in the context of a retrofit HEV, and the ECMS cost function is integrated into the MPC framework, which is successfully implemented in a Model-In-the-Loop (MIL) platform by execution under suitable driving cycles. For this model-based approach, a retrofit HEV plant model is developed using parameters acquired from an actual running retrofitted HEV having rule-based control in its supervisory controller ECU. Further, the acquired performance data of this vehicle provide a benchmark against the performances of MPC-based energy management strategies, one using a speed set-point error–based cost function and the other using the ECMS cost function, in MIL. Finally, comparative results and relevant analysis are presented to realize the energy-saving benefits and challenges of the proposed controller.