基于TEI@I-IOWA高速公路交通量预测研究

发布时间：2018-10-05 14:25

【摘要】：高速公路具有高能、高效、快速通达等优点,是地域或者地区之间相互联接的重要快速通道,能够促进沿线经济区域的繁荣发展,在国民经济发展过程中扮演着重要的角色。但是,高速公路也具有占地多、投资大、建设周期长,资金筹措困难等缺点,建设前期需要做大量的可行性论证,确定合理的建设规模才能发挥最大的经济效益。高速公路交通量大小是确定高速公路建设规模的最主要因素。无论是新建还是改扩建高速公路都需要作交通量需求预测分析,因此,能够准确预测高速公路交通量变化趋势能够为高速公路建设提供决策依据。本文重点研究高速公路交通量预测方法,以期能够实现对高速公路交通量的精准预测。高速公路交通量的变化受到多方面因素的影响。一个或者多个因素的变动都有可能引起交通量朝不确定的方向变化。因此,高速公路交通量预测是一项复杂的系统工作,需要不断发掘交通量发展变化规律以及不断改进预测算法,从而到达准确预测交通量变化的目的。本文在研究现有高速公路交通量预测技术方法的基础上,引入TEI@I方法论,基于其‘先分解后集成’的思想,将高速公路交通量预测分解成线性、非线性和时间序列三个部分,针对每部分的特点选取对应的方法进行单项预测,最后应用诱导有序加权平均(IOWA)算子集成单项预测结果,构建起高速公路交通量组合预测模型,并将组合预测模型应用于实例分析。实践研究结果表明,无论是从预测精度、整体误差,还是预测稳定性能上,基于TEI@I方法论的高速公路交通量组合预测模型的预测效果明显优于单项预测方法,能够较准确预测交通量的变化趋势。
[Abstract]:Expressway has the advantages of high energy, high efficiency, fast access and so on. It is an important link between regions and regions. It can promote the prosperity and development of economic areas along the route and play an important role in the process of national economic development. However, the expressway also has many shortcomings, such as large area of land, large investment, long construction cycle, difficulty in raising funds, and so on. In the early stage of construction, a large number of feasibility arguments should be made, and a reasonable construction scale can be determined to bring into full play the greatest economic benefits. The traffic volume of expressway is the most important factor to determine the scale of expressway construction. It is necessary to forecast and analyze the traffic volume demand no matter whether it is a new expressway or an extension expressway. Therefore, it is necessary to accurately predict the trend of highway traffic volume change and provide a basis for the decision making of highway construction. This paper focuses on the study of highway traffic volume prediction method, in order to achieve accurate highway traffic volume prediction. The change of expressway traffic volume is affected by many factors. Changes in one or more factors can cause traffic volume to change in an uncertain direction. Therefore, the expressway traffic volume prediction is a complicated system work, which needs to explore the changing rules of traffic volume development and improve the prediction algorithm constantly, so as to achieve the purpose of accurate prediction of traffic volume change. On the basis of studying the existing technical methods of expressway traffic volume prediction, this paper introduces the TEI@I methodology, and based on its idea of "decomposing first and then integrating", decomposes the expressway traffic volume prediction into three parts: linear, nonlinear and time series. According to the characteristics of each part, the corresponding method is selected to carry out the single prediction. Finally, the combined forecasting model of expressway traffic volume is constructed by integrating the single prediction results with the induced ordered weighted average (IOWA) operator. The combined prediction model is applied to the case analysis. The results of practical research show that the forecasting effect of the combined forecasting model of expressway traffic volume based on TEI@I methodology is obviously better than that of single forecasting method in terms of prediction accuracy, overall error and prediction stability. Can accurately predict the changing trend of traffic volume.
【学位授予单位】：湖北工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.14

【参考文献】