A kind of traffic data prediction method and system

Abstract

The invention discloses a traffic speed prediction method, which is characterized in that the method comprises: acquiring a data set, the data set including road network information and vehicle speed information; Spatial correlation and output the vehicle speed feature matrix V(K) of each road section; use the time convolution network to learn the time law of V(K) and output the time series feature matrix, use the multi-head self-attention mechanism to learn the coupling relationship between feature matrices and Output Through the present invention, the graph attention mechanism and temporal convolutional network are used to extract the spatial and temporal dependencies respectively, and the multi-head self-attention mechanism is introduced to extract the spatio-temporal coupling effect. The multi-head self-attention mechanism can learn from different representations at different positions. Spatial information, which helps to capture the temporal information of traffic flow and achieve higher prediction accuracy.

Description

technical field [0001] The invention is suitable for the technical field of intelligent traffic and data processing, and in particular relates to a traffic data prediction method and system. Background technique [0002] Traffic condition forecasting is an essential component in traffic modeling, operations and management. Unreasonable dredging when the traffic flow is large may lead to problems such as traffic jams and traffic accidents. High-accuracy traffic speed prediction plays an important role in intelligent transportation system (ITS) applications, such as alleviating urban traffic problems, improving urban transportation efficiency, helping travelers to do better route planning and departure timing, dynamic traffic signals Optimization and ramp control, etc. The methods of traffic speed prediction can be mainly divided into two categories: classical statistical models and machine learning methods. [0003] Most of the classical statistical models are based on som...

AI Technical Summary

Problems solved by technology

Use the cyclic neural network for timing prediction, but the cyclic neural network may forget some important information when learning a long sequence, and it cannot be calculated in parallel

Images