摘 要

近年来全球各国对化石资源短缺和环境污染严重关切，而电动汽车承载着环保节能理念，在我国发展势头迅猛。相对于蓬勃发展的电动汽车产业，电动汽车配套设施的发展显得比较滞后，尤其是充电设施的布局，在很大程度上限制了电动汽车的保有量。为了使电动汽车的充电需求得到满足，减少充电站闲置，为电动汽车发展提供基础，本文研究新规划集中式充电站的选址布局，对充电站的选址方法进行研究。

本文将城市中电动汽车充电需求简化为点需求，采用改进的p-中值选址方法进行研究，对于单个用户从已知的道路网中提取出道路拓扑图，并利用Dijkstra算法求得到各候选充电站位置的最短路以及实际道路距离，再由不同土地级别的时间权重对路段速度进行折算，得到时间最短路。以行驶时间最小以及排队时间和成本的线性加权和最小为目标函数，结合排队论的多服务台M/M/c理论，并设置约束条件，在无法采用精确计算法时引入贪婪取走启发式算法对模型求解。

本研究的创新在于将响应时间通过速度分级下的路段行驶时间和排队等待充电时间表示，不同于以往研究中仅考虑充电需求点到充电站的绝对距离，更加具有实际价值。应用排队论原理，将电动汽车等待充电的行为抽象为多服务台模型，在一定程度上影响充电桩中充电桩的数量设置。采用一致矩阵法找出线性加权系数，对两因素的重要程度进行对比。本研究意在对实践起到一定的指导作用，优化社会资源的配置，促进电动汽车产业可持续发展。

关键词：电动汽车；充电站；p-中值模型；排队论；贪婪取走算法

Abstract

In recent years, countries all over the world have been seriously concerned about the shortage of fossil resources and environmental pollution, while electric vehicles have carried the concept of environmental protection and energy conservation, and their development momentum has been swift and violent in China. Compared with the vigorous development of the electric vehicle industry, the development of electric vehicle supporting facilities appears to be lagging behind, and the layout of charging facilities, in particular, has limited the quantity of electric vehicles to a large extent. In order to meet the charging requirements of electric vehicles and reduce the idleness of charging stations, which provides the basis for the development of electric vehicles, this paper studies the layout of the location of centralized charging stations in the new plan and studies the method of selecting the location of charging stations.

This article simplifies the charging requirements of electric vehicles in cities into point requirements, adopts an improved p-median location method, and extracts the road topology from a known road network for individual users, and Dijkstra algorithm is used to find the nearest candidate charging station location and the actual road distance, and then the road speed is converted by the time weights of different land levels, and the shortest time is obtained. The objective function is to minimize the travel time and the linear weighted sum of the queuing time and cost, combined with the queuing theory M/M/c theory of multiple service desks, and setting constraint conditions, the greedy take heuristic algorithm is introduced to solve the model when the exact calculation method cannot be adopted.

The innovation of this research is to express the response time by the road section travel time under the speed classification and the queued waiting charging time, which is different from the previous research only considering the absolute distance of charging demand point to charging station, which has more practical value. Applying the theory of queuing theory, the behavior of waiting for charging electric cars is abstracted into a multi-service desk model, which affects the quantity of charging piles in the charging station to some extent. The uniform matrix method was used to find out the linear weighting coefficients, and the importance of the two factors was compared. This study aims to play a guiding role in practice, optimize the allocation of social resources, and promote the sustainable development of the electric vehicle industry.

Key Words: electric vehicle (EV); charging station; p-median model; queuing theory; greedy algorithm

目录

摘 要 I

Abstract II

第1章 绪论 1

1.1 研究背景及意义 1

1.2 国内外充电站选址研究现状 2

1.2.1国外研究现状 2

1.2.2国内研究现状 3

1.3国内外充电设施发展现状 4

1.3.1国外充电设施建设状况 4

1.3.2国内充电设施建设状况 4

1.3.3充电站的结构 5

1.3.4充电站对电力系统产生的影响 5

1.4 研究内容、方法及技术路线 6

1.4.1 研究内容及方法 6

1.4.2 研究技术路线 7

第2章 相关原理及基础理论 8

2.1电动汽车充电设施简介 8

2.1.1电动汽车 8

2.2.2电动汽车充电服务设施 8

2.2选址理论介绍 9

2.2.1p-中值问题 10

2.2.2p-中心问题 11

2.2.3覆盖问题 12

2.2.4截流问题 13

2.3排队论 14

2.3.1排队过程的一般表示 14

2.3.2排队系统的组成和特征 15

2.3.3排队模型分类 15

2.3.4到达间隔时间分布和服务时间分布 16

2.4图与最短路问题 17

2.4.1图的基本概念 17

2.4.2道路拓扑图 18

2.4.3最短路问题 18

2.5选址模型常用求解算法 19

2.5.1精确算法 19

2.5.2启发式算法 19

2.6多目标线性加权和法 19

第3章 基于p-中值模型的选址研究 21

3.1问题描述 21

3.2模型构建前准备 22

3.2.1模型假设 22

3.2.2模型参数和符号 22

3.3数学模型 23

3.3.1Dijkstra算法 23

3.3.2电动汽车充电行为排队模型 24

3.3.3改进的p-中值模型 26

3.3.4约束条件分析 27

3.4相关参数 27

3.4.1充电桩的建设成本 27

3.4.2充电站中充电桩数量上限 28

3.4.3需求点到候选站点的行驶时间 28

3.4.4线性加权系数 28

3.5模型求解 30

3.5.1模型求解思路 30

3.5.2算法原理及步骤 30

第4章 案例分析 31

4.1城市简介及电动汽车发展概况 31

4.2规划区域及道路拓扑图 31

4.2.1选择规划区域 31

4.2.2道路拓扑图的提取 32

4.2.3需求点位置和候选站点位置 33

4.3模型当中相关参数的取值 34

4.3.1充电桩的建设成本 34

4.3.2充电站中充电桩数量上限 35

4.3.3需求点到候选站点的行驶时间 35

4.3.4线性加权系数 40

4.3.5其他参数取值 42

4.4模型求解结果 42

4.4.1贪婪取走启发式算法的贪婪准则 43

4.4.2贪婪取走启发式算法 43

4.4.3运行结果 43

4.5灵敏度分析 48

4.5.1不同平均充电时间对结果的影响 48

4.5.2不同线性加权系数对结果的影响 48

第5章 结论与展望 52

5.1本文研究结论及成果 52

5.2研究创新点 53

5.3研究展望 53

参考文献 55

致 谢 58

附录A.Matlab程序代码 59

附录A1 Dijkstra算法 59

附录A2 输入函数和求解模型 61

绪论