Prediction method of urban congestion propagation mode based on cyclic autoregression model

Abstract

The invention relates to a prediction method of an urban congestion propagation mode based on a cyclic autoregression model. The method comprises the following specific steps: S1 generating a road section connection relation graph N by utilizing a topological connection relation of an urban road network; S2 selecting historical data in a long time period, and constructing a congestion propagationtime sequence knowledge graph G based on the road section connection relation graph N; and S3 training a cyclic autoregression model, learning a congestion propagation mode in the congestion propagation time sequence knowledge graph G, and performing prediction. According to the method, by constructing the corresponding time sequence knowledge graph, space-time prediction can be carried out for different types of congestion propagation modes in a road network.

Description

technical field [0001] The invention belongs to the field of intelligent traffic engineering, and relates to a method for predicting urban congestion propagation patterns based on a cyclic autoregressive model. Background technique [0002] Real-time or near-real-time predictions for the spatio-temporal propagation patterns of congestion in urban road networks will significantly improve the control, operation, and management capabilities of urban traffic. However, the traffic flow of the urban road network presents a strong randomness and temporal-spatial correlation, and the spatio-temporal prediction of the congestion propagation mode has become a major problem in the industry. [0003] Traditional traffic models attempt to model congestion itself. However, there are two major problems in the traditional model: First, the traditional model often judges the road sections with frequent congestion based on traffic flow or driving trajectory, and grades the importance of inte...

AI Technical Summary

Problems solved by technology

However, there are two major problems in the traditional model: First, the traditional model often judges the road sections with frequent congestion based on traffic flow or driving trajectory, and grades the importance of intersections or road sections. It focuses more on the occurrence of congestion than the propagation of congestion Secondly, the traditional method often only considers the congestion between adjacent road sections, or the congestion between road sections without any distance restrictions, and it is relatively simple to consider the intensity and spatial influence of congestion propagation

However, these knowledge graph models lack sufficient semantic richness (expressing different types of congestion propagation modes), and because they are static graphs (entities and relationships are fixed), they lack the dynamic expression ability of congestion propagation and cannot express congestion propagation Spatio-temporal evolution ability

Images