The Process Optimization of Train Operation with Time Constraint Based on Improved Multi-objective Memetic Algorithm

The process optimization of train operation is a sophisticated multi-objective optimization problem, especially in the condition of timing constraints. Due to the fact that the multi-objective optimization model of train operation process with time constraint is difficult to be solved precisely in the limited optimization time, the process optimization of train operation with time constraint based on multi-objective memetic algorithm (MMA) is proposed by using weight adaptive particle swarm optimization (PSO) and fuzzy logic in this paper. In order to enhance the global searching performance of the proposed algorithm, based on the analysis of the influence of inertia weight PSO global convergence, an improved PSO is presented to obtain the better optimization results. The shortcoming of the conventional PSO can be improved by optimizing the inertia weight and the evolution algebra relation curve. Meanwhile, for accurately obtaining a better solution in the neighborhood of the solution, a local search algorithm is developed by using a fuzzy decision based on the fuzzy logic of driving experience The advantages of the proposed method are summarized as follows: Firstly, a better convergence effect can be obtained by the global search algorithm. Secondly, by combining the fuzzy logic of driving experience and the fuzzy decision, the local optimization effect can be improved Finally, simulation results are provided to show the efficacy of the proposed method