基于支持向量机的高速公路物流量预测研究

发布时间：2018-02-28 23:09

本文关键词： 支持向量机高速公路物流量预测区域划分　出处：《昆明理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着世界经济环境的不断改善,经济全球化浪潮的不断冲击,我国在世界环境中扮演的角色也变得越来越重要。经济的快速发展带动了我国产业体系的不断完善,新兴的物流产业也开始了蓬勃发展,物流的概念和未来的发展已经悄然地成为全球经济体制发展的重要专注点和一种新的经济增速方式,是一个国家和地方经济综合实力的重要体现。物流技术的应用已经融入到了各个领域,各个政府和地区的部门都积极投入到物流基础设施的建设中去,公路运输是物流产业得以快速发展的主要方式。因此,高速公路物流量的研究也随之展开。本文的内容主要是针对高速公路的物流量预测进行研究。首先,通过对现阶段高速公路物流量预测方法进行了系统的梳理,总结不同预测模型存在的优缺点,并利用相关数据进行了模型精确度的对比。通过实验结果可以得出,在针对物流量的预测方面,支持向量机预测模型具有更高的准确性。其次,详细介绍了支持向量机预测模型的原理和步骤,描述了其数学算法的过程。结合物流量预测的具体情况,提出了构建基于支持向量机的物流量预测模型,对区域的物流量进行预测分析。再次,通过对高速公路所在区域的物流量特点研究之后,构建了物流量的区域划分模型,从而对某一条高速公路的物流量进行精确预测,并利用车流量的数据对区域划分模型进行调节修正。由物流量预测模型和区域划分模型两部分组合成新的基于支持机的高速公路物流量预测模型,提出了减少预测误差、提高预测结果可信度的方法。最后以昆明市嵩昆高速公路为例,利用构建好的高速公路物流量预测模型进行了实例研究,对嵩昆高速公路的客运量和货运量进行预测,得到未来的物流量数据。由于目前针对某一条高速公路物流量的研究还没有有效的展开。通过本文的研究,提出基于支持向量机的高速公路物流量预测模型,旨在为将来人们在关于高速公路的物流量预测方面提供一个新的思路。
[Abstract]:With the continuous improvement of the world economic environment and the constant impact of the wave of economic globalization, the role of our country in the world environment has become more and more important. The emerging logistics industry has also begun to flourish. The concept of logistics and its future development have quietly become an important focus point for the development of the global economic system and a new way of economic growth. It is an important embodiment of the comprehensive strength of a country and a local economy. The application of logistics technology has been integrated into various fields, and various government and regional departments have actively invested in the construction of logistics infrastructure. Highway transportation is the main way for the logistics industry to develop rapidly. Therefore, the research of expressway logistics flow is also carried out. This paper summarizes the advantages and disadvantages of different forecasting models, and compares the accuracy of the models with the relevant data. Support vector machine (SVM) forecasting model is more accurate in the aspect of material flow prediction. Secondly, the principle and steps of SVM prediction model are introduced in detail. This paper describes the process of its mathematical algorithm. Combined with the specific situation of the logistics prediction, the paper puts forward a logistics forecasting model based on support vector machine to predict and analyze the region's mass flow. Based on the study of the characteristics of the regional mass flow of the expressway, a regional division model of the flow of goods is constructed, which can accurately predict the flow of a certain expressway. Using the traffic flow data to adjust and modify the area division model, the paper combines the two parts of the material flow prediction model and the area partition model to form a new highway logistics forecasting model based on the support machine, and proposes to reduce the prediction error. Finally, taking Song-Kun Expressway of Kunming City as an example, the paper makes a case study of the constructed expressway material flow forecasting model, and forecasts the passenger and freight volume of Songkun-Kunming Expressway. To get the future material flow data. Because the current research on a certain highway logistics has not been carried out effectively. Through the research of this paper, a support vector machine based expressway mass flow prediction model is proposed. The purpose of this paper is to provide a new way for people to predict the mass flow of freeway in the future.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U492.31

【参考文献】