The invention discloses a short-term traffic flow prediction method based on a Spearman-LSTM model, and particularly relates to a region-level traffic flow prediction method. The method comprises the following steps: S1, collecting traffic flow characteristic and speed characteristic data, and preprocessing the collected data; s2, Spearman correlation coefficients of different regional road sections are calculated, the degree of correlation with data of a predicted road section is determined, and historical traffic data is divided into a training set and a test set; s3, constructing a Spearman-LSTM short-term traffic flow prediction model which comprises a long short-term memory network (LSTM) used for time sequence feature extraction and an attention mechanism used for spatial feature extraction; and S4, training the Spearman-LSTM network by using the training set, and testing the prediction precision of the model by using the test set. According to the method, the Spearman adjacent matrix is constructed, and the attention mechanism and the long short-term memory network (LSTM) are input, so that the model prediction precision is effectively improved.
本发明公开了一种基于Spearman‑LSTM模型的短时交通流预测方法,尤其为一种区域级交通流预测方法。该方法步骤如下:S1、采集交通流量特征和速度特征数据,并对所采集的数据进行预处理;S2、计算不同区域路段的Spearman相关系数,确定与预测路段之间数据的相关程度,并将历史交通数据划分为训练集和测试集;S3、构建Spearman‑LSTM短时交通流预测模型,包括用于对时间序列特征提取的长短期记忆网络(LSTM),以及用于空间特征提取的注意力机制;S4、采用训练集训练Spearman‑LSTM网络,再用测试集测试该模型的预测精度。本发明通过构建Spearman邻接矩阵并输入注意力机制和长短期记忆网络(LSTM),有效提高了模型预测精度。
Short-term traffic flow prediction method based on Spearman-LSTM model
一种基于Spearman-LSTM模型的短时交通流预测方法
2024-06-25
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Short-Term Traffic Flow Prediction: Using LSTM
IEEE | 2020
|Short-term traffic flow prediction method based on Conv1D-LSTM model
European Patent Office | 2023
|Short-term traffic flow prediction method based on improved LSTM
European Patent Office | 2021
|