Fine-Tuning a Constrained Track Alignment for Faster Travel on High-Speed Railways
针对铁路提速中轨道线形微调易产生不可行方案的问题，构建三维边界走廊约束，提出粒子群-摆动迭代法优化列车运行速度以缩短旅行时间，并通过实际案例验证。
The alignment fine-tuning search space for operating railways’ speed enhancement is constrained within a structurally adjustable range, defined by an extremely narrow 3D corridor. Existing methods for alignment refinement typically adopt a trial-and-error approach, adjusting design variables such as points of intersection and curve radii. However, these methods often produce a large number of infeasible alignments that violate the boundary constraints. To overcome this limitation, a three-dimensional boundary corridor (3D-BC) is constructed based on the adjustable ranges of measurement points. The alignment fine-tuning process is then confined within this 3D-BC. Specifically, a fine-tuning model is proposed for optimizing the trains’ operating speed to minimize travel time. In this model, the alignment design variables are initially represented by sets of horizontal and vertical arrays, from which the trains’ travel time between two endpoints is computed and designated as the optimization objective for enhancing railway speed. Geometric constraints, maximum allowable travel time, and the 3D-BC constraint are also incorporated in the model. To solve this model, a Particle Swarm Optimization (PSO)-Swing iteration method is developed. During the PSO evolution, measurement points are directly optimized, and their performance is evaluated by fitting them into an alignment with minimal deviation using the Swing iteration algorithm. In addition to initially confining the measurement points within the 3D-BC, the corridor is abstracted into a potential field that guides the measurement points away from the 3D-BC, when they are too near it. Finally, the effectiveness of the proposed method is demonstrated through a real-world case study.