부산대학교 도서관

The traditional genetic algorithm has been widely used in energy-saving optimization of train operation, but due to the uncertainty of the evolution direction of it's population chromosomes, it has caused some problems such as a large amount of local computations, slow convergence speed, and low solution quality. To solve these problems, an improved genetic algorithm is put forward in this paper. In order to effectively reduce the traction energy consumption of high-speed trains in operation, a multi-mass point train model based on the train dynamics model and the train running time and speed as the constraints is established. The improved genetic algorithm is used to analyze the search condition of conversion point of the working conditions, and the traction-lazy mode is used to reduce the energy consumption of the trains. The improved genetic algorithm takes the minimum energy consumption of the train as the optimization goal, converts the driving safety, punctual parking and other constraints into penalty functions, adopts the strategy of proportional selection and elite retention in the selection process, and introduces the adaptive crossover rate and adaptive mutation rate, which enhances the local and global search ability of the algorithm. Simulation results show that the improved genetic algorithm is suitable for energy-saving operation optimization of high-speed trains, which effectively improves the convergence speed and search ability of the algorithm. Compared with traditional genetic algorithms and adaptive genetic algorithms, the optimization results are more energy-efficient.