Fuzzy Random Optimization For Train Operation In Emergency

ABSTRACT:The infrastructure of high-speed railway is extensively developed in China for the past several years. The network topology structure and operation mode of the railway are changing profoundly. The pattern of train operation in China is the one which takes the sections as the dispatching units. The theory system on network based dispatching has not been formed and the corresponding methods are rare, which seriously affects the capacity of the whole transport system. Railway natural disasters occur frequently, and there are occasional railway accidents railway safety public incidents in China. Capacity loss of high-speed railway that assumes the primary responsibility for passenger transport will affect the whole transport system. Furthermore, train operation in emergency involves diverse objects, complex relations and various uncertainties. It makes the railway management more difficultly. Thus, it is a key issue to achieve the train operation theory, method and strategy in emergency under the background of the rapid development of Chinese high-speed railway.In this paper, we define the scientific issue of train operation optimization in different emergency, and analyze the principles, strategies and processes in different emergency. Then, our research focus on the problem of Train Operation in Serious Emergency and attribute it to an emergency triggered, bi-level programming and fuzzy and random parameters included, multi-stage, closed-loop optimization process.We analyze the uncertain parameters of section capacity in emegency and establish the capacity calculation method based on fuzzy Markov chain. Then, the section capacity in emergency is expressed by fuzzy random variable. Considering train speed, line capacity, line length and other factors in emergency, we present a measure of capacity-related——Time Strength of Line Capacity and Time Density of Line Capacity. Rerouting path search algorithm is proposed based on the time reliability, and it obtains the rerouting path set which adapts to the actual operating environment.In additional, we propose a bi-level programming to handle line plan adjustment and timetable rescheduling problem in emergency. The top layer objective is to make an optimal dispatching plan with selected actions which including merging trains, cancelling trains and making detour to other railway lines. This will take advantage of network capabilities to complete the passenger transport. Given a specific dispatching plan, the second layer (timetable rescheduling) of the optimization model focuses on minimizing the total delay as well as the number of seriously impacted trains by taking the strategies of changing the running time at a section, the dwell time at a station and the train stopovers. Then we analyze the uncertain factors of the model and put forward to expectation-tolerance model and chance-tolerance model according to the preferences of decision makers. It is solved by Branch and Bound Method mixed Multi-direction Plant Growth Simulation Algorithm (DPGSA). This ensures the implementation of the principle of train operation.At last, we complete the cases studies of Beijing-Shanghai high speed railway. Based on the validity of bilevel programming and tolerance model, we realize the whole process of capacity evaluation in emergency, rerouting path searching, timetable adjustment with speed restriction, bi-level optimization with line interrupted and train return. The different optimal schemes, according to the preferences of decision makers, prove the theory and method of Train Operation in Emergency in this paper.