Design optimisation of catenaries for mixed trains at different speeds

This research focuses on optimising the catenary for mixed trains with varying operating speeds. A weighted average of the contact force standard deviations at different speeds was set as the objective function for the purpose of improving the electric current collection quality. The dynamic data during the train movement was obtained from the finite element method (FEM) simulations. To expedite the optimisation process, the modified particle swarm - simulated annealing (PSSA) algorithm was applied to optimise catenaries. The algorithm was improved by avoiding repetition of FEM simulations with any duplicated input parameters and replacing the worst-performing model before lowering the simulated annealing temperature, which facilitates a larger scope for identifying the global optimal solution. To assess the effectiveness of the structurally optimised design, a comparison was made between the best optimised system and the original system. The results indicate that the weighted average of the contact force standard deviations of the best optimised catenary design is reduced significantly by 55.1%, which also outperforms the original design when operating in a large speed range and operating with different types of pantographs, and exhibits reduced vertical displacement at supports as well.