基于深度学习的城市快速路交通流预测方法

Urban expressway traffic flow prediction method based on deep learning

交通流预测在路径引导、交通管控和交通信息服务等方面具有重要意义,已成为近年来的研究热点。为应对不同日期、道路情况和天气情况对交通流波动产生的影响,提出了一种基于时间聚类的交通流量预测模型TC-ConvLSTM。从微波检测器收集原始流量数据,清洗数据,去除异常点,使用K-shape聚类方法对交通流量进行时间聚类,并针对不同的簇使用卷积长短时记忆神经网络预测交通流量。与其他深度学习方法的对比实验表明:在考虑不同交通模式分类下进行交通流量预测,能够取得更高的精度。

Traffic flow prediction is of great significance in route guidance,traffic control,traffic information services,and has become a research hotspot in recent years.In order to deal with some influencing factors,such as different dates,road conditions,and weather conditions on traffic flow fluctuations,a traffic flow prediction algorithm TC-ConvLSTM based on time clustering is proposed.First,collect original traffic flow data from the microwave detector,clean the data and then remove abnormal points in the data.Second,use the K-shape clustering method to cluster the data in time.For different clusters,convolutional long and short-term memory neural network is used to predict traffic flow.Comparative experiments with other deep learning methods show that traffic flow prediction can reach higher precision while considering the classification of different traffic modes.