面向城市路网的短时交通参数预测方法研究

发布时间：2018-05-07 03:17

本文选题：交通参数预测 + 卡尔曼滤波　；参考：《吉林大学》2017年硕士论文

【摘要】：随着城市交通的不断发展,智能交通技术不断革新,城市交通问题的解决也迫在眉睫。城市交通诱导已成为解决大中城市的交通问题有效策略,在城市诱导中,需要对交通信息进行有效、实时预测,从而交通管理者能够根据交通信息分析下一时刻交通运行状况,进而提出有效的诱导方案。因此,交通参数预测在交通管理中的作用已不容忽视。本文针对交通问题现状,首先对数据采样间隔及西宁数据分析,找出合适的采样间隔及预测时间间隔;提出了基于灰色关联度分析的Kalman滤波算法进行交通参数预测,基于历史数据进行预测,结果拟合较好;考虑到城市中部分路网检测器尚未覆盖,或检测器损坏等问题,采用相关性分析的方法,针对无检测器路段历史数据获取困难的情况进行分析解决。本文从实际角度出发,采用大量交通数据进行实例分析、并对所提方法进行验证。验证结果表明本文方法能够有效的解决实际交通参数预测的问题,并且能够为交通诱导提供有效的数据支撑。
[Abstract]:With the development of urban traffic and the innovation of intelligent transportation technology, it is urgent to solve urban traffic problems. Urban traffic guidance has become an effective strategy to solve traffic problems in large and medium-sized cities. In the process of urban guidance, traffic information needs to be effectively predicted in real time, so that traffic managers can analyze traffic conditions at the next moment according to traffic information. Then an effective induction scheme is proposed. Therefore, the role of traffic parameter prediction in traffic management can not be ignored. In this paper, according to the current situation of traffic problems, we first analyze the data sampling interval and Xining data, find out the appropriate sampling interval and forecast time interval, and propose a Kalman filter algorithm based on grey correlation analysis to predict traffic parameters. Based on historical data, the results are well fitted, considering that some road network detectors in cities are not covered, or detectors are damaged, the method of correlation analysis is adopted. The problem of obtaining historical data without detector is analyzed and solved. In this paper, a large number of traffic data are used for practical analysis, and the proposed method is verified. The results show that the proposed method can effectively solve the problem of actual traffic parameter prediction and provide effective data support for traffic guidance.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.14

【相似文献】