论文部分内容阅读
针对城市道路拥堵问题的日益加剧的问题,智能化城市交通管理平台是缓解拥堵问题的有效方法,利用交通流大数据预测结果进行交通诱导,能够指导用户调整出行方案,有效缓解交通压力。研究了交通流大数据的分布式增量聚合方法,对海量交通流数据进行清洗统计,为交通流预测提供数据基础,基于交通流在路网中上下游路段的相关性分析,利用路口转弯率多阶分配将该相关性量化,构建基于路网相关性的空间权重矩阵,完成对于STARIMA模型的改进。通过应用试验证明,该方法能更准确的进行交通流预测,为交通诱导信息发布提供依据。
In view of the increasing problem of urban road congestion, the intelligent urban traffic management platform is an effective way to alleviate the congestion problem. Traffic forecasting based on the big data of traffic flow forecasting can guide the user to adjust the travel plan and relieve the traffic pressure effectively. Based on the analysis of the correlation between the upstream and downstream traffic flow in the road network and using the intersection turn rate Multi-stage allocation quantifies this correlation and constructs a spatial weight matrix based on road network correlation to complete the improvement of the STARIMA model. It is proved by application experiments that this method can predict traffic flow more accurately and provide basis for traffic guidance information release.