基于EEMD-GSVM的高速铁路短期客流预测

发布时间：2018-06-20 23:21

本文选题：经验模态分解 + 高速铁路　；参考：《北京交通大学》2012年硕士论文

【摘要】：近几年来,我国高速铁路事业发展迅速,中国已成为世界上高速铁路发展最快、系统技术最全、集成能力最强、运营里程最长、运营速度最高、在建规模最大的国家。然而,中国高速铁路的发展离不开基础研究,客流量是修建高速铁路的基础和依据,合理的组织客流是发挥高速铁路效益的关键,因此,分析高速铁路客流,研究精确合理的高速铁路短期客流预测及方法具有重要的理论价值和现实意义。本文首先从客流形成机制、行为特性以及组合预测三个角度综述了高速铁路客流预测的理论与方法,为短期预测模型的建立奠定理论基础；其次,应用统计与时间序列挖掘的方法从波动周期、振位、能量、模式混叠等角度对高速铁路客流的波动特性进行分析,应用经验模态分解(EMD)的方法对高速铁路客流波动信号进行模态分析；第三,应用改进经验模态分解(EEMD)模型,借助“隔离”预测的理念,构建融合灰色生成与支持向量机的EEMD-GSVM短期预测模型,并研究了模态重构与参数的优化的方案；第四,以武广高速铁路为研究对象,运用所构建的EEMD-GSVM短期预测模型对高速铁路短期客流预测进行了实证分析,预测结果取得了较高的精度,说明EEMD-GSVM模型有较好的适应性；最后总结了论文的研究成果,并提出了有待于继续深入研究的问题。
[Abstract]:In recent years, the high-speed railway industry in China has developed rapidly. China has become a country with the fastest development, the most complete system technology, the strongest integration ability, the longest mileage, the highest operating speed and the largest scale under construction in the world. However, the development of high-speed railway in China can not be separated from basic research. Passenger flow is the basis and basis for the construction of high-speed railway. Reasonable passenger flow organization is the key to bring the benefit of high-speed railway into full play. Therefore, the passenger flow of high-speed railway is analyzed. It is of great theoretical value and practical significance to study the accurate and reasonable prediction and method of short-term passenger flow on high-speed railway. In this paper, the theory and method of passenger flow prediction of high-speed railway are summarized from three aspects: the formation mechanism, behavior characteristics and combination forecasting of passenger flow, so as to lay a theoretical foundation for the establishment of short-term forecasting model. The fluctuation characteristics of passenger flow in high-speed railway are analyzed from the angles of wave period, vibration position, energy and mode aliasing by using the methods of statistics and time series mining. The method of empirical mode decomposition (EMD) is used to analyze the wave signals of passenger flow in high-speed railway. Thirdly, the improved empirical mode decomposition (EMD) model is applied to predict the passenger flow in high speed railway with the help of the idea of "isolation" prediction. The short term prediction model of EEMD-GSVM combined with grey generation and support vector machine is constructed, and the scheme of modal reconstruction and parameter optimization is studied. The short-term forecast model of EEMD-GSVM is used to analyze the short-term passenger flow forecast of high-speed railway, and the forecast result is high precision, which shows that the EEMD-GSVM model has good adaptability, and finally summarizes the research results of this paper. The problems to be further studied are put forward.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：U293.13

【参考文献】