Study On Optimization And Application Of Rail Transit Passenger Flow Prediction Model

Urban rail transit is a type of safe and efficient mode of passenger transports,meanwhile,it has the characteristics of long planning years,huge cost and so on.Therefore,the scientific passenger flow prediction before construction can make the rail transit be reasonably designed.At present,the four stage method is widely used in passenger flow prediction.Through the study,it is found that the traditional four stage method has obvious shortcomings in the calculation of travel generation and travel distribution volume.The results often fail to meet the precision of the passenger flow prediction in some urban cases that with strong characteristics.With the promulgation and implementation of "Chifeng city comprehensive transportation planning",the public transport priority development strategy is clearly put forward in the 13 th five-year plan,Chifeng city begins to make a plan for the construction of rail transit.The city is located in the temperate and semi-arid continental monsoon climate zone,which has long snowy winter and many droughts in spring,and the traffic status are not ideal.The construction of rail transit will bring a new opportunity for the development of the city,and the level of traffic service will also be greatly improved.The traditional four stage prediction model does not take into account that the improvement of travel service level has the impact of traffic volume.At the same time,there are great differences in the planning of various regions in this city,the traditional prediction model ignores the trip generation caused by the interaction between the traffic zones.In view of the above problems,the main content of passenger flow prediction is introduced in this paper,research status at home and abroad is expounded,and the shortcomings of travel generation forecasting model and travel distribution forecasting model are discussed.Then,the prediction model of travel generation is revised.This paper creates a new idea of generation forecasting,analyzes the influence factors of various traffic generation and adds error compensation parameters.In accordance with the nature of land use,traffic district is divided,traffic generation for different properties of land is calculated respectively,which is the amount of travel generation in the traffic zone.Traditional traffic distribution prediction models do not take into account the internal travel of the traffic zone.In this paper,the models for traffic volume prediction of traffic zone interior and between the traffic zone are established respectively,making up for the shortcomings of the traditional models.An example is given to demonstrate the advantages of the new trip generation prediction model.The new model is convenient to calculate,its results are more suitable for the actual situation and the compatibility is better,which can lay a solid foundation for the subsequent prediction stage.The new model is applied in the forecasting of Chifeng rail transit passenger flow and processed by the survey data from the current year(2015).Combining with the data of 2006 resident trip survey and other related data,it is repeatedly proofread to prove the traffic models calibrated can simulate the characteristics and laws of current resident travel.According to the new models,the system of urban rail transit passenger flow forecasting(CRTP)is designed based on VC++ language and MFC class library.Then it is applied to the passenger flow indexes forecasting of Chifeng rail transit construction planning network,choose the better benefit one as the recommended line network.Finally,the passenger flow indexes of the recommended network characteristic year are predicted further,and the traffic effect of the recommended network scheme is compared with the requirement of comprehensive traffic planning.