Prediction Of Bus Operation State And Scheduling Optimization Based On Digital Twin

In recent years,the passenger of bus has declined as a whole,because the bus system is complex and seriously disturbed by the factors of traffic flow,traffic lights and inequality of passenger flow,which leads to the problems of low punctuality rate,weak stability and low service efficiency.It is still difficult to predict the arrival time and generate the optimal scheduling strategy,so as to stabilize passenger’s travel time and improve the level of bus service.In this context,this thesis will design the digital twin bus system and its operation mechanism based on the digital twin theory,and then carry out the following key technology research:(1)Rolling prediction of bus arrival time.This thesis uses the real-time interactive characteristics of the data of the digital twin bus system,synchronizes the data of vehicle location and passenger into the virtual space.Based on LSTM,the rolling prediction model is constructed to realize the prediction of arrival time of multi station.The model has a good prediction effect through four groups of experimental data.(2)Bus scheduling optimization.This thesis overcomes the shortcoming that the classic traffic simulation software can not reflect the passenger flow,establishes a bus operation simulation model with high confidence based on the digital twin,and solves the multi-objective bus timetable optimization model by using genetic algorithm which evaluate by the simulation model.On the basis of timetable,the max-min ant system is used to solve the vehicle scheduling model.Through the arrangement of the optimal chains of trips to optimize the utilization efficiency of vehicles.Two case are given to verify the effectiveness of the above two optimization models.(3)The development of bus synthetical management system.The above models and algorithms guide the system requirements analysis,functional module division,development environment construction and database design.After that,we develop a bus synthetical management system.The system integrates data monitoring,abnormal diagnosis,state prediction,scheduling optimization and simulation evaluation,and provides support for realizing the intelligence and refinement of bus dynamic management and control.