A Deadbeat Predictive Adhesion Control Method With Generalized Optimal Creep Model for HST Considering Low Adhesion Condition

Abstract

Running high-speed trains (HSTs) at the optimal slip ratio can significantly improve energy efficiency and actively prevent wheel slip. However, the intrusion of a third medium (such as water, oil, snow, etc.) introduces random uncertainties in the wheel–rail system’s creep characteristics, making it challenging for existing methods to estimate and track the optimal slip rate accurately, especially under high-speed conditions with low adhesion. To address this, a deadbeat predictive adhesion control method with a generalized optimal creep model is proposed. First, a predictive bandwidth (PB) adhesion force observer is introduced to determine the wheel–rail adhesion coefficient quickly and accurately. Next, a generalized optimal creep model is developed, which incorporates the measured wheel–rail creep curve for running speeds ranging from 40 to 400 km/h under water medium conditions, enabling the identification of the optimal slip ratio. Building on the deadbeat control, a fast adhesion torque controller is then designed to track the optimal slip ratio. The hardware-in-the-loop (HIL) simulation results validate the effectiveness of the proposed method. Even when the train operates under high-speed low-adhesion conditions at 300 km/h, the proposed method achieves an adhesion utilization rate of 94.38%, which is a 12.5% improvement over the traditional PI adhesion control method.