基于遗传算法的交通流量组合预测研究  

Combined Prediction Research of City Traffic Flow Based On Genetic Algorithm

作　　者：宋曰聪[1] 胡伟[1] 张涛[1] 

机构地区：[1]绵阳师范学院计算机科学与工程系,绵阳四川621000

出　　处：《微计算机信息》2007年第29期55-56,166,共3页Control & Automation

基　　金：四川省重点科技项目(02GG006-036);绵阳师范学院科研基金支助项目

摘　　要：智能交通系统是目前世界上公认的解决城市交通拥堵问题的最佳措施,实时、准确的交通流量预测是智能交通系统的关键技术之一,也是实现智能交通诱导及控制的前提。本文首先对几种重要的交通流量预测模型的理论和优缺点进行了比较,分析了影响预测模型的因素,然后提出了一种基于遗传算法的流量组合预测方法,该方法利用遗传算法群体搜索的特点,组合各种算法,优化预测思路,充分发掘不同算法的差异优势,实践证明该思路是切实有效的。Intelligent Transportation System is the best measure to solve the urban traffic jam in the world; forecasting urban traffic network is the premise for developing urban intelligent transportation system. In this paper, some important forecasting models, including the theory and characteristic, are discussed, and the factors are discussed to influence the forecasting model. In the end ,combined prediction of city traffic flow based on genetic algorithm is given, using the characteristics of genetic algorithm＇s colony search, the new algorithm combines all kinds of algorithms, optimizes the prediction way of thinking , fully discovers the advantages of different algorithms ,and turns out to be practical and productive.

关 键 词：智能交通 遗传算法 组合预测 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]