Torsional Vibration Suppression During Mode Transition Process in a Parallel Hybrid Electric Vehicle Based on Multiple Model Predictive Control

The mode transition process from electric mode to hybrid electric mode may cause a deterioration in the comfort of the parallel hybrid electric vehicle. To tackle this issue, a coordinated control strategy considering torsional oscillation of mode transition are developed in this study. First, the powertrain topology and powertrain dynamic are introduced to build a model for the transition system, which provides an environmental basis for the subsequent verification of coordinated control strategy. Second, an extended kalman filter is applied to accurately estimate the transmission torque, which is applied as a feedback variable to build a multiple model predictive controller (mMPC), different predictive models are selected according to the clutch state, and the optimal control parameters of the engine, motor and clutch are obtained by solving the quadratic programming problem aimed at comfort, economy and rapidity. Finally, the simulation results and hardware-in-the-loop (HIL) tests show that vehicle jerk are suppressed, and the driving comfort can be improved accordingly. At the same time, the influence of the cost function, the weight coefficient, and the predictive time domain on the torsional vibration suppression effect during the mode transition process are explored in the mMPC strategy design, and the solution time is controlled within 10ms, which meet the real-time requirements. Thus, this study provides a reference for vibration control of the parallel hybrid powertrains.