Road section traffic flow prediction method based on time convolution neural network

Abstract

The invention discloses a road section traffic flow prediction method based on a time convolution neural network, and the method comprises the following steps: combining collected historical traffic flow data according to a specified time interval, carrying out the normalization processing, and dividing the normalized traffic flow data into a training data set, a verification data set and a test data set; establishing a traffic flow prediction method based on a time convolutional neural network, training a time convolutional neural network prediction model by using the training data set, screening out an optimal prediction model by using the verification data set, and performing traffic flow prediction on the test set by using the time convolutional neural network model; and comparing theprediction data with the actual data, and performing error analysis. Based on the time convolution neural network structure, the space-time correlation of the traffic flow is considered, the trainingtime is short, and the prediction precision is high.

Description

technical field [0001] The invention relates to the field of intelligent traffic systems, in particular to a method for predicting traffic flow in road sections based on temporal convolutional neural networks. Background technique [0002] With the continuous advancement of urbanization, intelligent transportation systems have gained more and more attention in urban governance. The traffic information prediction system is an important part of the intelligent transportation system. Accurate and timely traffic information can help us make better travel decisions, alleviate traffic congestion, reduce urban environmental pollution, and improve the happiness of urban living. With the increasing popularity of various sensors and cameras, we obtain more and more traffic operation data. How to mine and utilize the rich traffic big data to make accurate and timely traffic forecasts and provide support for urban traffic congestion management is an important topic. [0003] After sea...

AI Technical Summary

Problems solved by technology

Recurrent neural network (RNN) and its variant long short-term memory network (LSTM) have achieved good performance in traffic flow prediction, but such models can only capture the temporal correlation of traffic flow data, ignoring the space of traffic network Correlation, and the training of such models is time-consuming

At present, there is no traffic flow prediction method based on temporal convolutional neural network model

Images