公交电子站牌信息延迟的分析及预测研究

发布时间：2018-01-17 19:35

本文关键词：公交电子站牌信息延迟的分析及预测研究　出处：《重庆大学》2015年硕士论文　论文类型：学位论文

【摘要】：发展公共交通是缓解城市交通拥堵问题最有效的手段之一,电子站牌系统通过向出行者提供实时动态公交信息,能够提升公交出行吸引力。由于从获取基础信息到将动态公交信息发布到终端上,整个过程会产生各种信息延迟,或多或少降低了动态公交信息的可靠性,影响了电子站牌使用效果,尤其是当信息出现长时间延迟后,影响更为明显。为了减少信息延迟带来的影响,提升公交电子站牌信息服务水平,对各种信息延迟特征进行分析,并把握信息延迟规律具有十分重要的现实意义。目前国内外研究中对公交电子站牌中的信息延迟问题关注有限,缺乏对不同信息延迟的具体刻画和分析。因此,论文结合信息的传递和转换过程对信息延迟问题进行了研究,通过划分信息延迟类型,全面分析了不同信息延迟的规律特点及其影响,在此基础上有针对性的建立了延迟预测模型。论文主要内容包括:①结合信息延迟产生的环节和特点,划分并定义了信息延迟的类型。根据不同类型延迟的构成,具体分析了各类信息延迟的特征以及延迟对电子站牌使用效果的影响,进而使对信息延迟问题的研究更具针对性。②从时间和空间两个方面重点分析了GPS数据延迟的规律和特征,为后续GPS数据延迟预测模型的建立奠定了基础。在时间上,主要通过对延迟数据时间序列分段,研究了GPS数据总体延迟的时段变化特性。在空间上,基于K均值聚类算法提取了不同路段GPS数据延迟时间的隐性特征向量,分析了不同路段GPS数据的延迟特征,通过对数据延迟时间的不确定性程度进行量化,比较了不同路段GPS数据延迟特征的相似性。③针对具有不同延迟特征的路段,基于非固定平滑系数的一次指数平滑法和灰色理论分别建立了组合预测模型,采用了变权重的线性组合方式,实现了对路段GPS数据延迟时间的预测,最后结合实例分析,验证了该组合预测方法的可行性和有效性。通过上述研究可知,在各种信息延迟中,GPS数据延迟是影响公交到站信息时效性和可靠性的关键。GPS数据延迟不仅具有时段差异性,并且从不同路段传回的GPS数据也具有不同的延迟特征,部分区域传回的GPS数据延迟时间会存在较大的不确定性。根据不同路段传回的GPS数据的延迟特点,论文对GPS数据延迟进行了进一步预测研究,从而有效把握了GPS数据延迟的变化趋势。
[Abstract]:The development of public transport is one of the most effective means to alleviate the problem of urban traffic congestion. Electronic stop board system provides real-time dynamic public transport information to travelers. From obtaining basic information to publishing dynamic bus information to the terminal, the whole process will produce a variety of information delays, more or less reduce the reliability of dynamic bus information. In order to reduce the effect of information delay, improve the information service level of public transport electronic stop sign, especially when the information is delayed for a long time, the effect is more obvious. It is very important to analyze the characteristics of information delay and grasp the law of information delay. At present, there is limited attention to the problem of information delay in public transport electronic stop signs in domestic and foreign research. Lack of specific description and analysis of different information delay. Therefore, this paper studies the problem of information delay by dividing the type of information delay with the process of information transfer and transformation. On the basis of analyzing the characteristics of different information delay and its influence, the delay prediction model is established. The main contents of this paper include the link and characteristics of the generation of information delay combined with 1: 1. The types of information delay are divided and defined. According to the structure of different types of delay, the characteristics of information delay and the effect of delay on the use of electronic station signs are analyzed in detail. So that the study of information delay problem is more targeted from the time and space aspects of the analysis of GPS data delay law and characteristics. It lays a foundation for the establishment of the delay prediction model of GPS data. In time, the temporal variation characteristics of the total delay of GPS data are studied by segmenting the time series of the delay data. Based on K-means clustering algorithm, the hidden feature vectors of delay time of GPS data in different sections are extracted, and the delay features of GPS data in different sections are analyzed. By quantifying the uncertainty of data delay time, the similarity of GPS data delay features of different sections is compared. 3. 3. Based on the exponential smoothing method with unfixed smoothing coefficient and grey theory, the combined prediction model is established, and the linear combination of variable weights is adopted to predict the delay time of GPS data. Finally, the feasibility and effectiveness of the combined prediction method are verified by a case study. Through the above research, we can see that in all kinds of information delay. GPS data delay is the key to affect the timeliness and reliability of bus arrival information. GPS data delay not only has time difference, but also has different delay characteristics of GPS data returned from different sections. The delay time of GPS data returned from some regions will be uncertain. According to the delay characteristics of GPS data returned from different sections, this paper makes further research on GPS data delay prediction. Therefore, the trend of GPS data delay is effectively grasped.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495

【相似文献】