Research On Decision-making In Urban Rail Transit Energy Management Based On Case-based Reasoning

In recent years, the proportion of rail transportation in the public transport sector growing between the world’s cities. With the global energy situation is worsening, energy consumption of rail transportation issues become the focus of attention by academia. Namely, how to ensure the capacity and driving under the security situation, reducing the energy consumption of rail transport effectively. Existing research has focused on energy-saving technologies improving and rail transportation key equipment selection, line select and train marshalling. These methods need large hardware and pre-wiring system, have long implementation cycle. Based on train operation diagram adjust to optimize energy management, need less hardware investment, low-risk, is a feasible method.Train Graph impact by many complex factors. Multi-lines, different environmental conditions and traffic conditions corresponding to different diagram, but there is a running figure is the lowest energy consumption of the best run chart. Solving about optimal operation of each particular case is a solvable but obscure problem. Our research group has accumulated a lot of the best run chart under different conditions by preliminary studies. Got the case database correspond to the train status and optimum energy consumption figure. In this paper, we proposed to use of case-based reasoning based on experience, through retrieval of similar cases, find the closest similar case quickly, to obtain the best solution to solve the approximate figure of emergency or unexpected situations run under the solution.In this paper, we combed the literature about case-based reasoning within domestic and foreign, expounded the basic theory of case-based reasoning summarized the processes, applications and development direction of case-based reasoning. Based on the summarizing of energy consumption of transport factors, obtained the tree structure about rail transportation energy issues. Using relational database constructed rail transportation energy case base, laid the foundation for the subsequent rail case retrieval and reuse of energy.Different category properties have different effects on the determination of similar cases, especially for energy optimization problem based diagram. For the case when the attribute weights to determine the current limitations of the Lord, their objective weighting method this paper presents a comprehensive weight trivalent Crossover method for determining (Trivalent Cross Synthetic Weight, referred TSW method), the first under expert scoring using AHP given the subjective weights coj; then weights adjusted by entropy method, that full consideration of subjective factors basis, according to rights case property features informative amount to get the corrected weight ωj’, and then based on the coefficient of variation method, based on the attribute data values the weights obtained discrimination ωi, final weight value coj’and coi synthesized multiplying the weight of the final consolidated ωc. On this basis, in order to consider the case between rail transport properties of the correlation energy, an improved gray correlation degree of similarity calculation method (Modified Gray Correlation, referred to as the MGC method), the above right to get integrated with the traditional weight gray correlation algorithm integration, energy scenarios to complete the search. Contrast algorithm analysis of experimental data shows the integrated application of these new methods to improve the quality of search results.Achieved the energy consumption of rail traffic management decision-making prototype system Based on Case-based reasoning, function and operation of the system are described. Finally, the work of this paper are summarized and discussed.