Optimal control of heavy haul train on steep downward slope

摘要

In this paper, we investigate the optimal control problem of heavy haul train on steep downward slope. Aiming to reduce train's maintenance cost and, most importantly, to ensure operation safety, a multi-stage dynamic programming model is formulated with the minimized in-train forces criteria. To solve the proposed model, this paper introduces a Q-learning algorithm considering operational requirements and the characteristics of air braking. As the size of a standard lookup table increases exponentially, we improve the computational performance and solution quality by maintaining a bounded lookup table. The computational results demonstrate the effectiveness and performance of the proposed model and algorithm.