Passenger Flow Short-time Prediction And Marshalling Optimization Of Urban Rail Transit

With the acceleration of China’s urbanization process, the urban population is growing rapidly, traffic pressure is increasing and the city roads are crowded, which has brought great inconvenience to people’s normal travel, the urban transportation problem has become increasingly prominent. To solve this problem, developing urban rail transit is a good way to deal with urban traffic congestion.In these days, China’s urban rail transit construction is in an unprecedented period, but the rapid development has also brought lots of problems, urban rail transit passenger flow forecasting is inaccurate and the formation is inappropriate. The conventional passenger flow forecasting inaccuracy makes its design based on which operators marshalling inappropriate, resulting in this stage of urban rail transit crowded or shipped wasted, thereby causing the increase in operating costs of urban rail transit.To solve this problem, on the basis of existing research, this paper summed up the urban rail transit has the characters of cyclical, balanced and short-term volatility and change, to construct a short-term based on gray forecasting model and neural network model of urban rail transit passenger combination predictive models. Neural network model to fix part of the gray model residuals, as well as use of gray model is easy to find out the advantages of the data variation to correct the prediction error of the neural network model. For short-term prediction of urban rail transit sectional passenger has a certain rationality and reference, which can be used as the basis of optimizing the design of urban rail transit operators marshalling.Reference to the principle of intelligent traffic lights, the short-term prediction of urban rail transit operators marshalling design is much closer to the actual passenger flow. Using the historical section passenger of last week as a training sample, apply IGNN to make short-term prediction, the next cycle can be obtained by section passenger to meet the a passenger changing trends. On this basis the operators marshalling design, urban rail transit can adapt to traffic changes in real time, to meet the changing passenger demand. Historical passenger will be classified in accordance with the actual passenger operations plan update fixes passenger as the next cycle of the design basis of the short-term prediction and operational mashalling. Urban rail transit capacity can be improved to a certain extent by mashalling optimization based on short-term prediction of urban rail transit operators to enhance operational efficiency and reduce operating costs.