Dynamic OD Estimation Of Passenger Flow In Urban Rail Transit Network

OD dynamic estimation of passenger flow is one of the key technologies to realize the dynamic management and control of urban rail transit(URT).In order to solve the problem of which is mostly used in road traffic OD estimation haven't considered that the rail traffic transect flow is difficult to collect,so the existing OD dynamic estimation method has low estimation accuracy when applied in the field of URT.Focusing on the dynamic OD estimation model of URT network passenger flow,based on the URT historical data of Automatic Fare Collection.For the model of OD dynamic estimation of URT passenger flow,the method is combined with the characteristics of URT network structure and passenger flow information collection features,and expounds the dynamic OD estimation process of URT passenger flow,and then the key problem of OD dynamic estimation is defined.the state space model of dynamic OD estimation of URT passenger flow is developed,and the kalman filtering method is used,based on the dynamic flow relationship between OD flow and collection of passenger flow passing in and out station.In addition,by considering the constraint conditions that the state variables of the model should satisfy,kalman filtering method under the constraint condition is proposed to correct the estimation result of the standard kalman filtering method.For the performance evaluation of dynamic OD estimation model of URT passenger flow,mean relative error and root mean square error are used to test the accuracy of kalman filtering method which is developed.The results of the constrained kalman filter passenger flow OD dynamic estimation accuracy show that:the model has good estimation performance under the condition of different passenger characteristic day and time interval.In comparison with constrained kalman filtering method and unconstrained kalman filtering method of estimation performance,the estimation accuracy of constrained kalman filtering method is obviously improved,and the error of constrained kalman filtering method is less than 20 percentage,so constrained kalman filtering method proposed is reasonable and effective.In addition,constrained kalman filtering method proposed has good performance on the estimation of large passenger demand of OD,and it has general performance on the estimation of small passenger demand of OD,but because of its small passenger demand,therefore,it will not have a significant effect on the overall estimation.