物联网环境下基于改进蚁群的大城市公交快速响应、运能优化与评价研究

发布时间：2018-02-24 08:26

本文关键词： 城市公交系统公交流量预测公交线网优化优化效能评价　出处：《合肥工业大学》2014年博士论文　论文类型：学位论文

【摘要】：随着城市经济、规模的不断发展和人民生活水平的不断提高,我国大中城市的交通出行压力在不断上升。作为城市系统的重要组成部分,城市公共交通系统为城市系统内人员、物质的流动提供必不可少的物质条件,城市公交系统内交通需求与供给的不平衡将会给城市经济社会的发展带来非常大的影响。当前,优先发展城市公共交通系统、促进出行需求的多样性已经成为广大学者、政府部门公认的解决大中城市交通问题的最有效途径。因此,传统公交线网优化时仅仅考虑地面公交线路的布设,忽视其他公共交通工具的做法存在很大的局限性,单一的考虑地面公交线网的布设已经难以满足城市交通出行多样化的要求。在实际的公交线网规划时,需要综合考虑城市公交系统内地面交通、轨道交通等多方面的规划,从整体上发挥不同类型公交工具的优势,提高城市公交系统的服务效能。本文的研究主要围绕城市公共汽车交通网络规划的关键理论、技术展开,以学术意义和实用价值并重为原则,在城市公交网络流量预测、城市公交线网优化方法研究、公交线网优化设计模型与算法研究、公交线网优化效能评估研究等方面取得了一定的成果。本文的主要内容如下： (1)公共交通网络是城市发展的基础,本文在深入分析国内外研究现状的基础之上,确定了对合肥市城市公交系统的资源配置与优化应当坚持以局部调整与整体统筹的并举的原则。通过分析城市公交线网交通流量的变化,有效的掌握相应的交通出行规律,从而有利于根据实际情况制定合理的公交优化方案。本文以公交系统内站点的客流量变化为研究对象,通过深入分析其变化规律,针对不同类型的交通流量数据,分别提出了基于经验模态分解的时间序列预测模型和基于经验模态分解的支持向量机模型,并明确了不同模型下的对于公交流量序列预测的预测适用范围。 (2)城市公交线网是城市公交系统的基本构成和核心。本文针对城市公交线网的特点,尝试以蚁群优化算法进行公交线网优化研究。通过分析了传统蚁群算法的特点和不足,重点提出了基于信息素递减、半动态候选列表的蚁群算法,并提出了dijkstra-蚁群混合算法用于相应的公交线网优化分析,所提算法解决了传统蚁群算法的不足,并提高了蚁群算法运用于公交线路优化的求解效能。 (3)城市公交线路优化绩效评估对于公交线路布设效能的判定至关重要。通过深入分析国内外公交线网评价方法的优点及不足,同时针对公交线路布设不同阶段的特点,提取了相应的评价指标体系,上述评价指标体系可有效的适用于公交线路前期布设及后期运营效能的定量评估。提出了基于分层聚类的主成分分析法运用于公交线路优化效能评价。 (4)城市公交线网优化的研究并不是相互分割的部分,需要将公交规划的各个流程合理的融入整个公交线网优化体系当中,才能有效提高公交系统的服务效能。城市公交需求的变化是一个动态发展的过程,我国大城市的发展速度很快,公交规划不仅需要考虑长期的公交发展需求,还应当考虑较短时间范围内的公交需求变化。本文以合肥市一环内公交线优化为例,深入分析了一环内道路交通状况及轨道交通建设情况,提出了基于dijkstra-蚁群混合的公交优化算法,得到了基于距离最短和基于客流量最大的城市公交系统线网优化模型,同时对公交路线优化的前期数据采集、优化问题求解、后期备选方案评价等过程进行了研究。在提出的城市公交线网优化研究体系的基础之上,再以合肥市一环内线路为实例研究了区域内各级线路优化调整方法,并给出了相应备选方案集的效能评定。通过实例验证,证明了本文理论体系的有效性。
[Abstract]:With the continuous development of city economy, the scale and the improvement of people's living standards, China's large and medium-sized city traffic pressure on the rise. As an important part of city system, city public transportation system for the city personnel within the system, an indispensable material condition to provide the material flow, the imbalance of traffic demand and supply the city bus system will have great effect on the economic and social development of the city. Currently, giving priority to the development of city public transportation system, promote the diversity of travel demand has become the majority of scholars, the most effective way to solve the traffic problems in large and medium-sized city government departments recognized. Therefore, only consider the layout of the ground bus lines of traditional bus lines network optimization, ignoring the other public transport approach has great limitations, consider the ground transit network design has been hard to full The foot of city traffic in transit. The diversified requirements of the actual network planning, need to consider the city bus system in ground transportation, rail transportation and other aspects of the planning, play different types of transportation advantage as a whole, improve the service efficiency of city bus system. The key theory, this study mainly focuses on the city bus traffic network planning techniques, the academic significance and practical value both for the principle, network traffic prediction in city bus, city bus line network optimization method research, research design model and algorithm optimization of public transit network, bus network optimization efficiency evaluation research and achieved certain results. The main contents of this paper are as follows:
(1) public transport network is the foundation of city development, based on in-depth analysis of the research status at home and abroad, to determine the configuration and optimization of Hefei city public transportation system resources should adhere to the principles of local adjustment and overall planning. By analyzing the changes of traffic flow of city bus line network, effective control traffic rules corresponding to the bus, to formulate the reasonable optimization scheme according to the actual situation. Based on the change of passenger flow in site bus system as the research object, through in-depth analysis of the variation of the needle on the traffic flow data of different types, are proposed. Time series empirical mode decomposition prediction model and empirical mode decomposition the support vector machine model based on clear and different model for bus flow sequence prediction scope.
(2) city transit network is the basic and core of the city bus system. Based on the characteristics of city public transit network, try to study the ant colony optimization algorithm to optimize the transit network. Through the analysis of the characteristics of the traditional ant colony algorithm and the insufficiency, puts forward the ant colony algorithm based on pheromone declining, semi dynamic candidates list dijkstra-, and puts forward the ant colony algorithm is used to optimize the analysis of transit network corresponding to the proposed algorithm to solve the shortcomings of traditional ant colony algorithm, and improved ant colony algorithm applied to solve the performance optimization of bus lines.
(3) city bus route optimization performance evaluation is essential for the bus line layout efficiency determination. Through in-depth analysis of advantages of evaluation methods at home and abroad of transit network and the insufficiency, at the same time for the bus line layout features of different stages, extract the corresponding evaluation index system, the evaluation index system can be effectively applied to the bus lines the layout and post operational efficiency of quantitative evaluation is proposed. The principal component analysis hierarchical clustering method is used to evaluate the efficiency based on the optimization of bus lines.
(4) and of the city public transportation wire optimization part is not separate, need to process the traffic planning reasonable bus network optimization into the whole system, we can effectively improve the service efficiency of public transportation system. The change of city bus demand is a dynamic development process, the development speed of big city in China soon, not only need to consider the public transit planning long-term development needs, should also consider the bus demand change short time range. This paper takes the Hefei city bus line optimization within a ring as an example, in-depth analysis of the situation of road traffic and rail transit construction of a ring, the bus hybrid optimization algorithm dijkstra- based on the ant colony, has been based on the shortest distance and the largest traffic system of city bus line network optimization model based on the previous data acquisition bus route optimization, optimization problem solving, The alternative evaluation process was studied. On the basis of proposed city bus network optimization research system, and then study the adjustment method of optimization at all levels of lines within the region within a ring line in Hefei city as an example, and gives the evaluation of the effectiveness of the corresponding options set. Through the example, show the validity of this the theory of the system.

【学位授予单位】：合肥工业大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：U491.17

【参考文献】