Train Speed Trajectory Optimization using Dynamic Programming with speed modes decomposition

Abstract

When applying a dynamic programming algorithm to train speed trajectory optimization, there is a problem of too many discrete points which leading to dimension disaster. Different from using uniform discretization of the time and space by previous research, In this paper, a train operation model is proposed based on conditions (speed limit and slope change) and described in the network diagram. Armed with this model, through the pre-planning of the line, the optimization model of train operation is established, and the dynamic programming algorithm is utilized to find the global optimal value of train traction energy consumption. Taking Beijing Yizhuang line as a simulation case, the validity of the proposed model is verified by comparing common optimization algorithms. The results demonstrate that the proposed model can effectively reduce the calculation time of the dynamic programming algorithm by 95.06%, and has better optimization effect. The energy-saving effect reaches 6.0806%.