【作者】 于建玲； 商朋见； 关积珍；

【Author】 Yu Jianling1 Shang Pengjian2 Guang Jizhen1 (1.Beijing Stone Intelligent Transportation System Co., LTD, Beijing 100081, China; 2. Beijing Jiaotong University, Beijing 100044, China)

【机构】 北京四通智能交通系统集成有限公司； 北京交通大学；

【摘要】 本文运用相空间重构方法对交通流进行短时预测,首先根据重构展开法确定时间延迟、根据Cao-方法计算嵌入维数;然后在相空间重构局域方法的基础上,对交通流预测方法进行改进,给出预测模型。通过实证分析,结果表明改进后的预测模型达到较好的预测精度。更多还原

【Abstract】 In this paper, the phase space reconstruction is applied to the prediction of short-term traffic flow. First, we define the time delay according to the method of reconstruction-expansion and calculate the embedding dimension according to the method of Cao. Then on the base of the phase space reconstruction local-region method, we improve the method of prediction in the traffic flow. Meanwhile a new model is introduced to predict traffic flow. Case study using real data from RTMS proves the validity of the modified method.更多还原