Optimising the trajectories of multiple interacting trains to maximise energy efficiency is a difficult, but highly desirable, problem to solve. A bespoke genetic algorithm has been developed for the multi-train trajectory optimisation problem and used to seek a near-optimal set of control point distances for multiple trains, such that a weighted sum of the time and energy objectives is minimised. Genetic operators tailored to the problem are developed including a new mutation operation and the insertion and deletion pairs of control points during the reproduction process. Compared with published results, the new GA was shown to increase the quality of solutions found by an average of 27.6% and increase consistency by a factor of 28. This allows more precise control over the relative priority given to achieving time targets or increasing energy efficiency.
Multi-train trajectory optimisation to maximise rail network energy efficiency under travel-time constraints
2016-05-01
18 pages
Article (Journal)
Electronic Resource
English