一种基于灰色马尔科夫的大客流实时预测模型  

A grey Markov based on large passenger flow real-time prediction model

作　　者：杨军[1,2] 侯忠生[1] 

机构地区：[1]北京交通大学电子信息工程学院,北京100044 [2]北京市轨道交通指挥中心,北京100101

出　　处：《北京交通大学学报》2013年第2期119-123,128,共6页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家科技支撑计划项目资助(2011BAG01B01;2011BAG01B02);国家自然科学基金重点项目资助(60834001)

摘　　要：为分析和处理大型活动期间轨道交通大客流的预测预警问题,本文构建了一种基于灰色马尔科夫的大客流实时预测模型.以各类大型活动的历史OD客流数据为基础,利用灰色预测算法对客流数据建立灰色模型,然后建立马尔科夫修正模型,最后利用预测误差对灰色预测结果进行修正得到最终的预测大客流值.实验采用北京轨道交通OD客流数据和标准评价方法,实时预测五棵松体育馆举办的CBA决赛对五棵松地铁站产生的出站大客流,预测结果表明模型是有效的,对真实的大客流预测具有较好的效果.In order to analyze and figure out the problem of metro system large passenger flow prediction and warning during the large-scale activities, this paper introduces a novel grey Markov based real-time prediction model for large passengers. Based on the historical origination-destination （OD） passenger data during different large-scale events, this model establishes the grey model for passenger time-series data by grey prediction method at first, and then constructs the improved Markov model, and gray forecasting results are corrected by the prediction error to acquire the final predicting large passenger number at last. The experiments are based on Beijing metro system OD data and standard evaluation methods, and predict the real-time outbound large passenger number on Wukesong station affected by CBA final matches in Wukesong gymnasium. These results show that the presented model is valid and has good prediction performance for actual large passenger flow.

关 键 词：轨道交通 大客流 灰色模型 马尔科夫模型 

分 类 号：U231.92[交通运输工程—道路与铁道工程]