| With the development of the social economy,the urbanization process of our country is accelerated gradually.As an important travel tool for modern urban residents,the subway has the characteristics of large capacity,fast speed,high punctuality and small area.It can effectively relieve urban traffic pressure.However,due to the large volume of the MTR,its operating energy consumption is also large,which bring the enterprise enormous costs and energy pressure.Therefore,it is urgent to analyze the influence factors of energy consumption of metro line and study the energy-saving technology of subway.In the past 10 years,the research of this research group has found that,for a given metro line,the same hardware facilities are used.Given the same external conditions(flow conditions,fault points,etc.)the difference between the minimum and maximum energy consumption is close to 2.3%.In other words,there is an "optimal operation diagram" for a certain external condition of the subway,and the Tsim rail transit electrical simulation system developed by our research team calculates such an optimal operation diagram.The laboratory needs 35-120 hours.Such a calculation speed,is not meet the metro's real-time needs.Therefore,how to get the "lowest energy consumption" optimal operation diagram quickly and efficiently under the new external conditions is the main objective of this paper.In this paper,case-based reasoning technology is used to calculate the energy consumption data based on the given operation diagram of urban metro and the experimental data of Tsim rail transit electric simulation system based on other feasible charts.A case-base of energy-saving metro diagram is constructed.Each case in the case base is composed of a triad:the external conditions of the subway operation,the lowest energy consumption of the external conditions,and the lowest energy consumption.This paper analyzes the factors that affect the energy consumption of the subway,and uses the framework knowledge representation to express the case,and designs the multi-level index structure and case structure of the case database.Aiming at the problems existing in the subjective weighting method,we propose a variable weight comprehensive method,and calculate the case attribute the weight value of the indicator.Aiming at the problem of large dimension and magnitude difference in attribute in case,an improved case method of variable weight nearest neighbor method is proposed.This method normalizes the case attribute value according to the linear function method.The paper achieves the prototype operating system of the case database,designs the overall structure and main functions of the system.In the prototype system,combined with the actual measured data of subway operation and the simulation data of Tsim system,the weight calculation method and the improved weight value proposed in this paper are experimentally verified.The results show that the Improved Weighted-value Nearest Neighbor Method is more accurate than the nearest neighbor method and the nearest neighbor method.The retrieval time to get the time of the optimal run diagram shortened to the original 1.4 ‰-2.2 ‰.After the target case performs the optimal run diagram corresponding to the most similar case,compared with the data measured by the MTR Corporation and the operation data without the optimal operation diagram,energy consumption decreases by 1.5%or less while the difference between the "enforced energy consumption" and the "theoretical optimal energy consumption" is within 5‰.Experiments show that in the case of 1200 base case,the optimal time is 3 to 16 minutes,which reduces the computation time of the optimal operation graph.The experimental results are satisfactory,and the aim of this paper is achieved. |
PDF Full Text Request