基于RTMS数据的区域机动车积累量分析和可视化

发布时间：2018-11-28 17:47

【摘要】：随着机动车数量的飞速增加,城市交通问题日益凸显,对人们的日常生活产生了巨大的影响。据统计,我国机动车保有量在2016年上半年已达到2.58亿,大量的机动车数量带来了严重的城市拥堵和停车困难等交通问题。为了解决交通问题,各个国家大力发展智能交通系统,通过实时监测交通系统,采集交通数据来分析、研究并制定交通措施。各国学者也利用采集的交通数据对交通系统展开了广泛的研究。但是,现有的研究主要关注于路网本身,对路段的运行状态进行研究。对于路段外,即停车问题的研究则较少。并且,已有的关于停车问题的研究也受到停车数据的限制,无法对没有数据的情况进行研究。而区域内的机动车积累量包含了路网中运行和区域中停靠的两部分机动车数量,涵盖了更加全面的交通信息。通过机动车积累量,我们既可以分析路网的运行状态,也可以推断区域的停车需求。因此,本文使用区域的机动车积累量进行研究。使用杭州市路段微波流量数据,提出并实现了基于微波数据的机动车积累量估计方法,并基于估计的机动车积累量对区域的交通状态进行分析,推算区域的停车需求。同时,也搭建了基于WEB端的可视化平台,对本文的交通数据和计算结果进行展示。首先,针对交通流量数据量大,传统工具效率低的问题,本文使用Spark分布式计算平台对流量数据进行预处理,极大地提高了处理效率。其次,通过对微波流量数据的分析,提出灰色模型和历史均值的加权方法来解决流量数据中记录缺失的问题,并使用实际微波数据与其他方法对比说明方法的性能。然后,根据路网数据对城市进行区域划分。针对区域机动车积累量的估计问题,提出利用进出区域的路段流量数据进行有向累加的基本方法来估计初步的机动车积累量结果。并进一步改进,提出基于区域划分的偏差分配补全方法来解决基本方法存在偏差的问题。并利用实际杭州市微波数据进行实例机动车积累量的估算,以此推断区域的停车需求,并对比实际真值验证本文方法的效果。最后,本文设计了可视化的方案并搭建了基于WEB端的可视化平台实现了交通数据的展示、分析功能。通过可视化平台来直观展示本文的微波数据,呈现区域机动车积累量的变化规律,分析不同数据之间的关系。
[Abstract]:With the rapid increase of the number of motor vehicles, urban traffic problems become increasingly prominent, which has a great impact on people's daily life. According to statistics, the number of motor vehicles in China has reached 258 million in the first half of 2016, a large number of motor vehicles have brought serious urban congestion and parking difficulties and other traffic problems. In order to solve the traffic problem, every country develops the intelligent transportation system vigorously, through monitoring the traffic system in real time, collecting the traffic data to analyze, studying and formulating the traffic measure. Scholars from all over the world also make use of the collected traffic data to carry out extensive research on the traffic system. However, the existing research mainly focuses on the road network itself. There are few studies on the parking problem outside the section. Moreover, the existing research on parking problem is limited by parking data. The amount of motor vehicle accumulation in the region includes the number of two parts of motor vehicles in the road network and in the region, which covers more comprehensive traffic information. Through the accumulation of motor vehicles, we can not only analyze the running state of the road network, but also infer the parking demand of the area. Therefore, this paper uses the area of motor vehicle accumulation to study. Based on the microwave flow data of Hangzhou section, a method of vehicle accumulation estimation based on microwave data is proposed and realized. The traffic state of the area is analyzed based on the estimated amount of vehicle accumulation, and the parking demand of the area is calculated. At the same time, the visualization platform based on WEB is also built to display the traffic data and calculation results. Firstly, aiming at the problem of large amount of traffic flow data and low efficiency of traditional tools, this paper uses the Spark distributed computing platform to preprocess the traffic data, which greatly improves the processing efficiency. Secondly, based on the analysis of microwave flow data, the grey model and the weighted method of historical mean are proposed to solve the problem of missing records in the flow data, and the performance of the method is illustrated by comparing the actual microwave data with other methods. Then, according to the road network data, the city is divided into regions. In view of the problem of estimating regional motor vehicle accumulation, a basic method of directional accumulation based on the flow data of road sections in and out of the region is proposed to estimate the preliminary results of motor vehicle accumulation. Furthermore, a new method based on region partition is proposed to solve the problem of deviation in the basic method. The actual microwave data of Hangzhou are used to estimate the accumulative amount of motor vehicle in order to infer the parking demand of the area and to verify the effectiveness of the method compared with the actual value. Finally, the visualization scheme is designed and the visualization platform based on WEB is built to display and analyze traffic data. The microwave data of this paper are displayed intuitively by visual platform, and the variation law of regional motor vehicle accumulation is presented, and the relationship between different data is analyzed.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.1

【参考文献】