When this study similarity algorithm, the variable weights Nearest Neighbor algorithm (VWTNN) and variable weights Nearest Neighbor algorithm case attributes were normalized

and improved variable weights nearest neighbor (modified VWTNN) were compared the study。 Finally, select the historical case base 100 cases as a case history library, enter the 30 detected cases, the two methods of calculation were obtained corresponding to the most similar case method; then in most similar cases run operation corresponding to FIG。 detection of cases, to give 30 cases were detected in two ways to match the most similar case diagram after running the composite score, obtained 90% of the cases detected with improved variable weights nearest neighbor (improved VWTNN) matching the most similar case, energy consumption, passenger satisfaction and transfer station dwell time limit under the corresponding composite score of cases detected nearest neighbor (VWTNN) composite score is equal to or higher than the value of the transformer right conclusions。

Keywords: Urban Rail Transit, Multi-objective decision, Energy Consumption , Case-Based Reasoning, Case Retrieval, Similarity, K-Nearest Neighbor

本 科 毕 业 论   文 第 I 页

目 次

1 引言 1

1。1 研究背景 1

1。2 研究意义 2

1。3 研究现状 2

1。3。1 轨道交通的研究现状 2

1。3。2 案例推理的研究现状 3

1。3。3 相似度的研究现状 3

1。3。4 研究趋势 4

1。4 本文的研究内容 4

2 案例相似度在轨道交通应用中的相关理论 5

2。1 案例推理的基本理论 5

2。1。1 案例推理的过程 5

2。1。2 案例推理系统的优点 5

2。2 层次分析法的基本理论 6

2。2。1 层次分析法的基本结构 6

2。2。2 层次分析法的步骤 6

2。3 相似度的基本理论 7

2。3。1 空间距离 7

2。3。2 相似度 8

3 轨道交通运行图多目标决策系统分析 10

3。1 多目标决策的发展历程 10

3。1。1 多目标决策的提出 10

3。1。2 多目标决策的发展