The invention relates to the field of congestion propagation analysis methods, in particular to an algorithm-improved congestion propagation analysis method. The method comprises the following steps of: a) processing data from an area by a congestion judgment module, dividing traffic states by calculating the saturation of an intersection, and calculating the saturation of each lane of the intersection by calculating a traffic demand index TI and the traffic capacity of the intersection so as to realize the division of the traffic states; b) integrating results obtained by the congestion judgment module by an item set generation module; and c) generating an FP-Tree frequent item set. According to the method, a frequent item set mining method of an Apriori algorithm is replaced by an FP-Tree algorithm, so that the operation efficiency of the algorithm is effectively improved. And data filtering is completed by adding spatial-temporal feature constraints, and invalid data in the strong association rules are filtered again by adding a lifting degree evaluation method. According to the method, the congestion trend can be quickly analyzed, and information is provided for urban congestion relief.
发明涉及拥堵传播分析方法领域,尤其是算法改进型拥堵传播分析方法。该方法的步骤为:a)拥堵判别模块对来自区域中的数据进行处理,通过计算路口饱和度进行交通状态的划分,通过计算交通需求指数TI和路口通行能力,从而计算出交叉路口的各车道饱和度,来实现交通状态的划分;b)项集生成模块将来自拥堵判别模块获得的结果进行整合;c)FP‑Tree频繁项集生成。本发明通过FP‑Tree算法代替了Apriori算法的频繁项集挖掘方法,有效提升了算法运行效率。通过添加时空特征约束完成数据过滤,并增加提升度评估方法再次过滤强关联规则中的无效数据。本申请可以快速分析拥堵趋势,为城市缓堵提供信息。
Algorithm improved congestion propagation analysis method
算法改进型拥堵传播分析方法
2022-03-11
Patent
Electronic Resource
Chinese
Modelling traffic congestion propagation based on data analysis method
British Library Conference Proceedings | 2022
|European Patent Office | 2024
|European Patent Office | 2016
|