Application of quantum-behaved characteristic particle swarm optimisation algorithm in multi-objective optimisation of urban rail train

Abstract

This paper combines operation control strategy and operation curve to address the problems of frequent switching and high energy consumption of traditional ATO control policies. Firstly, based on the Pareto principle, the objective optimisation model is established based on urban rail trains' punctuality and energy consumption. Then, a multi-objective quantum particle swarm optimisation (MOQPSO) algorithm with fewer control parameters is adopted. The Gaussian mutation operator and crowding distance sorting method are introduced to select the global optimal guidance particles better, and the optimisation effect of the conventional MOPSO algorithm is compared. Finally, the actual data of the Beijing Subway Yizhuang Line are used to verify the algorithm. Simulation results show that the MOQPSO algorithm has advantages in convergence, diversity, and optimisation. At the same time, different control strategies are compared, and the results show that the improved hybrid control strategy has a better optimisation effect on the longer line.