To address the inherent multidimensional conflicts among power continuity, powertrain durability, and fuel economy exacerbated by the massive inertia of hundred-ton hybrid electric mining trucks (HEMTs), a novel hierarchical decoupled control architecture is proposed. First, a dual-motor powertrain capable of power-interruption-free shifting is designed to overcome high-load dynamic constraints. To resolve the delayed gear decision-making under high-inertia loads, an intelligent optimal gear mapping scheme combining Dynamic Programming (DP) and Neural Networks (NN) is then developed, bridging offline global optimization with online real-time execution. Furthermore, a multi-objective adaptive energy management framework based on the Equivalent Consumption Minimization Strategy (ECMS) is established. Considering overall fuel economy, battery State of Charge (SOC) stability and engine start-stop frequency as optimization objectives, the equivalent factor, SOC penalty coefficient, and engine start-stop penalty coefficient are treated as optimization variables. The Artificial Tree (AT) algorithm is utilized to perform global parameter optimization under specific driving cycles. Simulation results show that, compared with the Rule-Based (RB) strategy, conventional ECMS, and Proportional-Integral ECMS (PI-ECMS), the proposed method achieves better SOC stability and improves fuel economy by 14.98%, 2.56%, and 0.90%, respectively. This improvement corresponds to a significant reduction of 122.45 kg of CO2 emissions per 100 km compared to the RB strategy. Meanwhile, engine start-stop frequency is reduced by 60.34%, 63.49%, and 65.15%, respectively, effectively balancing fuel economy and powertrain durability.
Zhang et al. (Mon,) studied this question.