ECMS-Based MPC Strategy for Multi-Mode Hybrid Electric Trucks

To address the adaptive energy management optimization challenge of a hybrid powertrain system consisting of an internal combustion engine and dual electric motors, an Equivalent Consumption Minimization Strategy (ECMS) is introduced into the Model Predictive Control (MPC) framework. This paper proposes an MPC-ECMS energy management strategy tailored for multi-mode hybrid electric trucks. First, a genetic algorithm is employed to optimize a radial basis function neural network, enhancing its predictive performance for short-term, high-precision vehicle speed forecasting based on historical data. Next, the ECMS is combined with MPC to solve for the minimum objective function value within the prediction horizon, thus obtaining the corresponding power distribution results. This enables real-time adaptive adjustment of the equivalent factor, with the prediction horizon rolling forward for continuous optimization. Simulation results demonstrate that the proposed control strategy outperforms the ECMS control strategy with a fixed optimal equivalent factor under the same operating conditions, achieving a 4.2% improvement in fuel consumption and reaching 93% of the fuel consumption performance of a dynamic programming-based approach.