基于IC卡数据的OD矩阵生成技术及其应用研究

发布时间：2018-08-15 12:00

【摘要】：城市居民公交出行数据是城市公交规划和运营调度的基础数据。如何获取居民公交出行数据以便采取措施应对不断增长的公交需求是当前城市公交建设亟待解决的问题。传统人工采集公共交通数据的方式不仅耗费大量的人力物力,采集到的数据也存在样本量少、精度低等不足。随着电子信息技术在城市公共交通系统中的应用,公交IC卡随之产生并得到广泛应用,为城市公共基础数据的采集提供了新途径。但是,大多数城市的公交IC卡票制是一票制,产生的公交IC卡数据不包含乘客的上下车站点信息,无法直接获取居民出行数据。因此,如何基于公交IC卡数据生成城市居民公交出行OD矩阵(Origin-Destination Matrix)显得极为重要。本文的主要工作内容和创新点为:(1)挖掘城市居民公交出行特征。通过对公交IC卡数据进行统计分析,挖掘了居民公交出行的时空特征,线路乘坐特征以及乘车行为特征。这些特征是基于IC卡数据生成居民出行OD矩阵的理论基础。(2)创新性地提出了一种生成特征上车站点的方法。该方法利用乘客每天早高峰首次乘坐的多条公交线路有共同站点这一特征,产生乘客早高峰时段首次乘车时的特征上车站点及特征线路集。通过判断IC卡刷卡记录中乘客的乘坐线路是否存在于其特征线路集中,即可确定乘客的上车站点。(3)基于乘客的特征上车站点和公交IC卡数据生成居民公交出行OD矩阵。利用特征上车站点判断部分乘客的上车站点,判断每趟公交车辆行驶方向,判断所有乘客的上车站点以及判断所有乘客的下车站点。然后,基于完整的IC卡数据生成居民公交出行OD矩阵。实验结果表明,本文的研究成果可用于判断IC卡数据中乘客的上下车站点,不仅可以用于推导居民出行OD矩阵,还可作为其他公交数据挖掘的基础数据。同时,本文将实验生成的公交出行OD矩阵映射为车辆出行OD矩阵,应用于微观交通仿真系统的实例中,得到了更加真实的仿真结果。
[Abstract]:Urban public transport travel data is the basic data of urban public transport planning and operation scheduling. How to obtain the data of public transport travel and take measures to cope with the increasing demand for public transport is an urgent problem to be solved in the construction of urban public transport. The traditional way of collecting public transportation data by hand not only consumes a lot of manpower and material resources, but also has some shortcomings, such as small sample size and low precision. With the application of electronic information technology in urban public transportation system, bus IC card is produced and widely used, which provides a new way for the collection of urban public basic data. However, the bus IC card system in most cities is one ticket system, and the generated IC card data do not contain the information of passengers' boarding and disembarkation stations, so they can not directly obtain the residents' travel data. Therefore, how to generate urban residents'OD matrix (Origin-Destination Matrix) based on IC card data is very important. The main contents and innovations of this paper are as follows: (1) excavating the characteristics of urban residents' public transport trip. Through the statistical analysis of IC card data of public transportation, this paper excavates the spatial and temporal characteristics of residents' public transport travel, the characteristics of line ride and the characteristics of ride behavior. These features are the theoretical basis for the generation of OD matrix for resident travel based on IC card data. (2) an innovative method for generating feature boarding stations is proposed. This method takes advantage of the common station of multiple bus lines that passengers take in for the first time in the morning rush hour of each day, and generates the characteristic boarding station and characteristic line set during the first time of passenger's early rush hour. By judging whether the passenger's riding line in the IC card credit card record exists in its characteristic line concentration, the passenger's boarding station can be determined. (3) based on the passenger's characteristic bus stop and bus IC card data, the residents' bus travel OD matrix is generated. The characteristic bus station is used to judge the boarding station of some passengers, to judge the direction of each bus vehicle, to judge the boarding station of all passengers and to judge the stop of all passengers. Then, the OD matrix is generated based on the IC card data. The experimental results show that the research results in this paper can be used to judge the stations of passengers in IC card data, not only to deduce the OD matrix of residents, but also as the basic data of other bus data mining. At the same time, the OD matrix generated by the experiment is mapped to the vehicle travel OD matrix, which is applied to the microscopic traffic simulation system, and the simulation results are more real.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.1;TN409

【参考文献】