Breaking the hierarchical silo: Structure–control co-optimization for hybrid mode switching under transient impact and fatigue constraints

Abstract

This paper presents a structure–control co-optimization method to mitigate planetary gear fatigue and extend system life in power-split hybrid commercial vehicles during EV–HEV mode switching, where transient jerk induces mechanical stress. First, key control parameters affecting jerk and engine start-up time are identified from an MPC-based switching strategy. A fatigue damage model is then built using ANSYS transient simulation and Miner’s rule, based on a full-cycle load spectrum. Structural parameters of the planetary gear and MPC weight coefficients are jointly treated as design variables. Surrogate models are developed through sampling simulations to link design variables to performance targets. A multi-objective optimization approach is applied to find the best parameter combination. Results show that after optimization, fatigue damage drops to 3.55 × 10⁻⁹, peak jerk is 8.19 m/s³, and engine start-up time is 0.34 s—all meeting design requirements. The results indicate that the co-optimization approach can better coordinate structural and control objectives than a conventional hierarchical design, particularly in managing trade-offs among durability, comfort, and responsiveness.