The invention relates to a short-term traffic flow prediction method based on a Conv1D-LSTM model. The method comprises the steps of S1, collecting traffic data and performing preprocessing; s2, calculating Pearson's correlation coefficients between different road sections, determining S adjacent road sections having significant spatial correlation with the predicted road section, and dividing historical traffic data into a training set and a test set; s3, establishing a Conv1D-LSTM short-term traffic flow prediction model which comprises a one-dimensional convolutional network for performing feature extraction on the input data and a long and short-term memory neural network for performing time sequence prediction; s4, performing iterative training on the Conv1D-LSTM short-term traffic flow prediction model by adopting the training set based on a time sequence back propagation algorithm; and S5, performing short-term traffic flow prediction on the test set by adopting the trained Conv1D-LSTM short-term traffic flow prediction model. Compared with the prior art, the method fully considers the dependency relationship of the traffic volume in two dimensions of time and space, and has the advantage of high prediction accuracy.
本发明涉及一种基于Conv1D‑LSTM模型的短时交通流预测方法,该方法包括:S1、采集交通数据并进行预处理;S2、计算不同路段之间的皮尔逊相关系数,确定与预测路段具有显著空间相关性的S个相邻路段,并将历史交通数据划分为训练集和测试集;S3、建立Conv1D‑LSTM短时交通流预测模型,包括用于对输入数据进行特征提取的一维卷积网络,以及用于进行时间序列预测的长短期记忆神经网络;S4、基于时序反向传播算法,采用训练集对Conv1D‑LSTM短时交通流预测模型进行迭代训练;S5、采用训练好的Conv1D‑LSTM短时交通流预测模型对测试集进行短时交通流预测。与现有技术相比,本发明充分考虑交通量在时间和空间两个维度上的依赖关系,具有预测准确性高的优点。
Short-term traffic flow prediction method based on Conv1D-LSTM model
一种基于Conv1D-LSTM模型的短时交通流预测方法
2023-07-14
Patent
Elektronische Ressource
Chinesisch
Short-term traffic flow prediction method based on improved LSTM
Europäisches Patentamt | 2021
|Expressway Short-Term Traffic Flow Prediction Based on CNN-LSTM
Springer Verlag | 2024
|Short-term traffic flow prediction method based on CASSA-LSTM algorithm
Europäisches Patentamt | 2022
|