摘要
设计并实现了基于多源交通数据的重大事件影响下城市交通客流预测实验平台。基于现有的交通预测方法基础上,提出对城市路网划分为多个子区域,基于区域内重大事件的举办通过融合天气、事件、时间等多源因素,分别对每个子区域内出租车的上下客情况,提出采用遗传算法优化BP神经网络进行出租车上下客数量预测。该实验平台选定纽约市麦迪逊广场花园附近1公里的区域为研究区域,应用文中的算法对该区域进行区域划分和客流预测,所提出的方法能够有效地提高预测精度。
This paper focuses on the traffic prediction methods,and forecasts traffic condition based on events.First,this paper proposes to partition the urban road network by dividing the entire urban area into multiple sub-areas.For,we integrate traffic condition,for example,weather,major events,and time.We design a novel platform to effectively predict the number of pick-up and drop-off by using the BP neural network optimized by the genetic algorithm to conduct prediction.The platform divided an area 1 km away from Madison Square Garden in New York City and predicted using the algorithm proposed in this paper.The prediction results demonstrate that the proposed algorithm can effectively improve the prediction accuracy.
作者
徐秀娟
张文轩
陈谆悦
赵小薇
许真珍
XU Xiu-juan;ZHANG Wen-xuan;CHEN Chun-yue;ZHAO Xiao-wei;XU Zhen-zhen(School of Software Technology,Dalian University of Technology,Dalian 116620,China)
出处
《实验室科学》
2020年第1期16-20,共5页
Laboratory Science
基金
国家自然科学基金项目(项目编号:61502069)
中央高校基本科研业务费(项目编号:DUT18JC39)。
关键词
路网分区
交通预测
遗传算法
BP神经网络
road network partition
traffic prediction
genetic algorithm
BP neural network