A traffic flow forecasting method based on LSTM_Attention network

Abstract

A traffic flow prediction method based on LSTM_Attention network comprises the following steps: (1) acquiring road traffic history data, dividing the data into training set and test set, and preprocessing the data; (2) constructing a layer of LSTM network, the number of hidden layer units is set according to the sequence length of samples in the training set data, a full-link layer with Softmax activation function is added, and finally a logistic regression layer is added as a prediction layer, the training set data is input into the network to obtain the prediction value, the prediction valueand the real value are input into the loss function, and the network model and the internal parameters are optimized by back propagation. (3) Inputting the test set data into the trained LSTM_Attention network to obtain the prediction data. The present invention accurately predicts future road traffic flow data.

Description

technical field [0001] The invention relates to a traffic flow prediction method based on an LSTM_Attention network. The invention belongs to the field of traffic flow prediction and relates to a fusion method of road traffic flow data and a neural network architecture. Background technique [0002] The rapid economic development and the continuous improvement of people's living standards have brought about the rapid growth of the total number of urban motor vehicles and the aggravation of road traffic load, and the problem of road congestion is becoming more and more serious. The distribution of vehicles in a transportation network is crucial to the circulation of the entire network. In order to achieve a reasonable distribution of vehicles, certain regulatory measures need to be taken to make the current and future vehicle distribution tend to be reasonable, so it is very important to accurately predict future traffic conditions. [0003] Traffic conditions can be measure...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem that the final output of a single LSTMs has an incomplete representation of the temporal characteristics of the input data in the prior art, the present invention introduces an attention mechanism to assign correlation weights to the output of each hidden layer unit of the LSTMs to obtain temporal characteristics. Finally, accurately predict the future road traffic flow data

Images