Forecast And Analysis Of Chongqing Rail Transit Passenger Flow Based On BP Neural Network

With the rapid development of China’s economy and the continuous improvement of science and technology in recent years,the construction of urban rail transit in China has gradually developed into a network operation.As the core part of the modern city traffic system,rail transport capacity,punctuality rate,to avoid congestion,average speed,traffic accident rate is low,energy and environmental protection features to fully demonstrate that the city rail transportation is becoming the preferred means of transport of the urban population,urban rail transit passenger flow has more characteristic and periodicity etc..Therefore,historical data,analysis of passenger rail transportation to explore the variation and modeling to predict the future traffic,reasonable arrangement of equipment operation and management departments of maintenance personnel,arranging appropriate timetable and improve Chongqing’s rail service quality has a very important guiding significance.Based on the passenger flow of Chongqing rail transit and the passenger flow distribution theory and passenger flow prediction theory,combined with Chongqing’s own characteristics,this paper studies and explores the passenger flow of Chongqing rail transit passenger flow.The main work and conclusions are as follows:1,we sampled and analyzed the passenger flow data of Chongqing rail transit,summarized the distribution law of passenger flow and the imbalance characteristics of time and space,illustrated the various imbalances,and described the relationship between the corresponding disequilibrium coefficients by mathematical formulas.2.Review the development process of passenger flow forecasting,compare the common passenger flow forecasting methods,summarize their respective characteristics,and analyze their applicability.Between the traditional short-term passenger flow forecast model cannot well show curve,which leads to BP neural network,introduces the working principle and process of nonlinear and uncertain problems on the edge,and on the basis of the actual demand for the corresponding parameters of the BP neural network setting method,the application of clustering analysis with the sample data,select more representative the sample collection,to make it meet the demand analysis,we make the prediction results more consistent with the actual traffic.3.All the data support from the introduction of all concepts in this paper are derived from the data statistics and combing of passenger flow in Chongqing rail transit network.Select the Chongqing rail transit line three,the largest passenger station in the Guanyin Bridge as the research object,the establishment of neural network time series regression model and neural network forecasting model,comparing the characteristics of two kinds of prediction methods,carry out analysis and use of a large amount of data processing software Matlab,realize the function of data flow prediction and error correction,construction of database of passenger flow finally,the realization of data accumulation and circulation of passenger flow prediction function.And within the allowable range of error,the correlation between the passenger flow and the objective factors of the future import and export station is summarized and summarized.