Multi-parameter collaborative optimization and performance evaluation of multimodal electro-hydraulic servo suspension system

Traditional single-mode suspensions are difficult to meet the compound demands of vehicles for high vibration reduction performance, operating reliability under varying road excitations, and low operating energy consumption. Existing studies on electro-hydraulic servo suspensions still have limitations in terms of unified multimodal coupling modeling, multi-parameter collaborative optimization, and the adaptability of conventional particle swarm optimization algorithms. This paper focuses on a multimodal electro-hydraulic servo suspension system and conducts a systematic study on system modeling, algorithm optimization, and performance evaluation. Based on a quarter-car model, the dynamic equations and hydraulic characteristic equations of the suspension system in passive, semi-active, and active modes are derived. An AMEsim-Simulink co-simulation model capable of representing the three operating modes through different valve-state configurations is constructed. To overcome the tendency of conventional particle swarm optimization to fall into local optima and the imbalance between global and local search capabilities, an improved particle swarm optimization algorithm integrating dynamic inertia weight and a random reflection wall strategy for out-of-bounds particles is proposed. A comprehensive objective function considering sprung mass acceleration, suspension working space, and tire dynamic load is constructed to achieve collaborative optimization of the damping apertures in passive mode and the PID gains in semi-active and active modes. Based on the filtered white noise method, random road excitations of Class A, B, and C were generated at a representative vehicle speed of 20 m/s. The optimization performance of PSO, GA, and IPSO and the vibration reduction performance of the optimized suspension system were evaluated under the considered road excitations. The results show that, under the considered Class A–C road excitations, the proposed IPSO achieves lower final fitness values and better convergence stability than conventional PSO and GA. Compared with the traditional single-damping passive suspension, the optimized passive, semi-active, and active modes generally improve the vibration reduction performance, while their advantages differ among road grades and evaluation indicators. These results provide a simulation basis for parameter tuning and subsequent mode-selection design of multimodal electro-hydraulic servo suspension systems.