Dynamic Performance and Rollover Stability Analysis of Hydrogen-Powered Heavy-Duty Vehicles Under Multi-Operating Conditions
Abstract
Hydrogen-powered heavy-duty vehicles (HHDVs) operating under multiple driving conditions are subjected to coupled longitudinal, vertical, and lateral dynamic excitations, which significantly affect their dynamic performance and rollover stability. To investigate these characteristics, a coupled vehicle dynamic model consisting of a vertical dynamic model and a yaw–roll dynamic model was established, and numerical simulations were conducted under multiple operating conditions. The effects of operating condition, road roughness, initial braking speed, and braking deceleration on ride comfort and dynamic tire load were systematically analyzed. Furthermore, rollover stability was evaluated under J-turn, Fishhook, and Double Lane Change (DLC) maneuvers using yaw rate, slip angle, lateral acceleration, and lateral load transfer ratio (LTR) as evaluation indices. The simulation results show that braking causes the greatest deterioration in ride comfort, with the peak human–seat vertical acceleration increasing by 33.10% compared with the constant-speed condition, while acceleration results in a 27.55% increase. Road roughness substantially affects both ride comfort and dynamic tire load. Under braking, the peak front and rear tire dynamic loads on a Class D road are approximately 3.1 and 2.9 times those on a Class B road, respectively. Increasing the initial braking speed intensifies dynamic responses, whereas increasing the braking deceleration effectively suppresses tire dynamic load fluctuations. Among the three steering maneuvers, the Fishhook maneuver exhibits the highest rollover propensity, with the maximum absolute LTR approaching 0.8. These simulation-based findings provide insights into chassis parameter optimization, vehicle dynamic performance evaluation, and rollover prevention of HHDVs under multiple operating conditions. The present study is limited by the lack of experimental validation of the developed dynamic models, and experimental or hardware-in-the-loop validation will be considered in future work.