基于随机森林模型的智慧高速公路交通事故预测研究与应用

Research and Application of Intelligent Expressway Accident Prediction Based on Random Forest

近几年来,高速公路事故发生率居高不下,事故后果严重、伤害性大,因此,众多学者开展高速公路事故预测研究。基于随机森林模型,以S45天黎高速公路五台至盂县段(K347+600-K362+600)为示范应用,提取历史数据中的事故样例与非事故样例进行对比,通过对历史交通流量和事故检测器数据进行分析,对高速公路上的事故风险进行实时预测,实现交通事故识别。

In recent years,the sustained high expressway accident rate has led to serious consequences and great injuries,so,lots of researchers have conducted studies of expressway accident prediction.In this paper,taking the section of Wutai to Yu County in S45 Tianli Expressway(K347+600-K362+600)as the demonstrating application,based on Random Forest,the accident examples were extracted from history data to compare to non-accident examples.Analyzing the historical traffic flow and accident detection data could predict the expressway accident risk in real time,realizing traffic accident recognition.