The invention discloses a traffic flow prediction method based on a neural network. The method comprises the steps: collecting historical traffic flow data of a prediction road section in set time, and forming a two-dimensional traffic flow matrix; standardizing the traffic flow data, and classifying the traffic flow data into a training set and a test set; designing an LSTM-FC neural network forpredicting the traffic flow, and training the LSTM-FC neural network by using the training set; testing the trained model by using the test set, evaluating a model error, and if the error is greater than a set value, retraining the model; and inputting the traffic flow data of a road section to be predicted in the first five days into the trained LSTM-FC neural network, and predicting the future traffic flow of the road section. According to the invention, the future traffic flow can be predicted according to the historical traffic flow data of the section, and the real-time performance and accuracy of traffic flow prediction are greatly improved.
本发明公开了一种基于神经网络的交通流预测方法,收集预测路段设定时间内的历史交通流数据,形成二维交通流矩阵;对交通流数据进行标准化处理,并分为训练集、测试集;设计预测交通流量的LSTM‑FC神经网络,并利用训练集对其进行训练;用测试集对训练结束后的模型进行测试,评估模型误差,若误差大于设定值,重新训练模型;将待预测路段前五天的交通流数据输入训练好的LSTM‑FC神经网络,预测路段未来交通流量。本发明能够根据断面的交通流量历史数据预测未来的交通流量,极大程度提升了交通流预测的实时性和准确性。
Traffic flow prediction method based on neural network
基于神经网络的交通流预测方法
2020-06-16
Patent
Elektronische Ressource
Chinesisch
Airport traffic flow prediction method based on graph neural network
Europäisches Patentamt | 2023
|Traffic flow prediction using neural network
IEEE | 2018
|Traffic flow prediction method based on Fourier-recursive neural network
Europäisches Patentamt | 2020
|Traffic flow prediction method based on EMD-GAN neural network structure
Europäisches Patentamt | 2021
|