The Research Of The Train Operation Scheme Optimization For Passenger Dedicated Lines Based On Passenger Oriented

Train operation scheme is the core of the train transport organization of passenger dedicated lines, the operating income from railway passenger transport can be improved by reasonable operation scheme. In the market economy, optimization of the passenger train operation scheme is also a major mode to improve competitiveness. Because there are many factors influencing the formulation of the passenger train operation scheme, with the increasing demand of passenger, an optimization of train operation scheme of passenger dedicated lines based on passenger oriented is put forward.Firstly, the domestic and foreign research status of train operation scheme and the main influence factors of the passenger flow volume have been analyzed in the paper. Considering the influence of various factors on the passenger flow, the BP neural network algorithm which has mature theory and the high fitting degree have been chosen, and the complex nonlinear mapping between the population, per capita income, GDP and the passenger flow volume have been realized, and the passenger flow of Wuhan-Guangzhou passenger dedicated line have been forecast.Secondly, the influencing factors, the formulating principles and the optimizing content of passenger train operation scheme for dedicated passenger lines were analyzed, the boundaries of assumptions for establishing the optimizing model was set, the determinate method and formulate principle of the originating and destination station were put forward, and the train operation log and the principle of distribution of passenger flow were determined. The benefit of railway passenger transport departments(maximum) and the passenger travel cost(minimum) were chosen as the target function, and the optimization model of passenger train operation scheme of dedicated passenger lines was established. The multi-objective programming model that established in this paper is a multi-objective nonlinear 0-1 integer programming model, it also could be summed up in 0-1 knapsack problem. The original multi-objective problems were changed into single objective programming problem through dimensionless process and the effective solutions of original problem was obtained after the calculation.Because the traditional genetic algorithm generally adopts the roulette wheel selection method, some excellent individuals will be eliminated due to the probability select in the process. Therefore, some improvements were been put forward in the process of effective solution selection, namely, the principle of eliminating inferior. The advantage of these improvements is that excellent individuals in the population will not be eliminated for the probability choice, and the best population can be got by calculation. The improved genetic algorithm was been used to solve the model by the C++ programming.Finally, the passenger flow OD table was predicted and train operation scheme was generated. In the total transport cost, passenger traveling time and revenue, the operation scheme compare with the existing operation scheme, and verify the effectiveness of the scheme.