Research On Optimization Of Energy-efficient Train Operation In Urban Rail Transit

With the process of urbanization,the development of urban rail transit in China has entered a new period.By the end of 2020,a total of 45 cities in China have opened and operated urban rail transit.The total length of the urban rail transit lines in China is 7978.19 kilometres.At the same time,the rapid expansion of urban rail transit has brought a lot of energy consumption which can not be ignored.The traction energy consumption is the main part of the total energy consumption in urban rail transit.Reducing the traction energy consumption has become a hot issue in academia and industry in recent years.The driving strategy which controls the speed of single train and the timetable which schedules the running time of all trains are two key factors to determine the traction energy consumption.Optimizing the train driving strategy and the timetable can effectively reduce the traction energy consumption and carbon emissions without changing the existing system structure and hardware equipment.The optimizations are significant to promote the development of green transportation and the construction of low-carbon economy in China.Optimizing the train driving strategy and the timetable need to make full use of the relevant functions of each subsystem in railway signalling system.This dissertation combines the principles of train safety protection in different block systems and the timetable scheduling characteristics under various operation modes to study the optimization problems about the energy-efficient train operation in urban rail transit.The optimization problems are proposed from three aspects: the optimization of single train driving strategy,the optimization of tracking train driving strategy and the integrated optimization of train dispatch and control.Exact algorithms are proposed for each optimization problems according to their characteristics in this dissertation.The major contributions of this dissertation are:(1)Aiming at the shortcomings of existing exact algorithms,such as low efficiency and difficult to satisfy the terminal conditions,an efficient dynamic programming approach is proposed to solve the problem of optimizing energy-efficient train driving strategy.In order to avoid the limitation of the “curse of dimensionality”,the algorithm searching space in this approach is constructed on a "speed-position" based two-dimensional discrete network.In order to satisfy the terminal condition of the punctual constraint,this approach introduces time resource constraints for each vertex in the discrete network,and builds an optimization model based on the shortest path problem with resource constraints.In order to improve the efficiency of the dynamic programming,this approach desgins a dominating operator to delete the dominated strategies generated in each optimization stage.Furthmore,this approach desgins a crowding operator to maintain the quantity and the diversity of the non-dominated strategies generated in each optimization stage.The results show that,compared with the existing exact algorithm,the efficiency of the proposed approach is significantly improved.The proposed approach results in better energy saving improvement and satisfy the punctual constraint.(2)In order to realize the energy-efficient train driving and improve the line capacity during the train dense tracking operation,this dissertation combines the principles of train safety protection in different block systems to study the problem of optimizing train headway and traction energy consumption under the most restricted area of main line.According to the characteristics that train tracking is limited under the station area in the moving block system,this dissertation takes both traction energy consumption and achievable minimum headway as indexes to optimize the train driving strategy.The ?-constraint method is introduced to bulid an optimization model based on the shortest path problem with resource constraints.An efficient dynamic programming approach is used to solve the optimal model.The optimization results reveal the equilibrium relationship between the objectives of energy-efficient train driving and line capacity improving in the moving block system.Furthermore,according to the characteristics of the train tracking limitation in the intermittent quasi-moving block system,this dissertation optimizes the train driving strategy under the most restricted block section of main line.(3)Aiming at the problem of low generation efficiency and bad quality of the energyefficient train strategies Pareto front in the integrated optimization of train dispatch and control,an efficient and exact method is proposed.In this method,energy saving and time saving are both taken as the optimization objectives.Combining with the dominating operator,a dynamic programming is carried out on a "speed-position" based two-dimensional discrete network to generate the Pareto front of the energy-efficient train strategies.Furthermore,this dissertation takes single train as object and builds an integrated optimization model of train dispatch and control under the conventional operation mode.According to the characteristic of the optimization model,an exact algorithm is proposed.The results show that,the generation efficiency of the Pareto front is significantly improved.Compared with the optimum train driving strategy under the original timetable,the integrated optimization results in 0.8 percent energy saving improvement.(4)According to the characteristic that the time and space orders of trains are changed after the overtaking,this dissertation studies the problem of integrated optimizing train dispatch and control under the express/local operation mode.For a given simple overtaking scenario,an integrated optimization model is constructed.In the model,the running times of express train and local train are taken as the decision variables.The minimum headways before and after the overtaking are taken as the constraints.Minimizing the traction energy consumption of both express train and local train is taken as the optimization objective.According to the characteristic of the optimization model,an exact algorithm is proposed.The results show that,compared with the optimum train driving strategy under the original timetable,the integrated optimization results in 1.1 percent energy saving improvement under the equilibrium departure mode,and further results in 0.1 percent energy saving improvement under the non-equilibrium departure mode.