Traffic flow prediction method of divergence convolution recurrent neural network based on space-time diagram

A technology of cyclic neural network and prediction method, which is applied in the field of traffic flow prediction of divergent convolutional cyclic neural network, can solve the flow process without considering the gradual divergence of traffic flow, and does not consider the spatiotemporal characteristics of traffic flow (time dependence and Spatial dependence and other issues, to achieve the effect of accurate prediction and high degree of fitting

Inactive Publication Date: 2019-11-22
CHONGQING CITY MANAGEMENT COLLEGE
View PDF7 Cites 42 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] ① The spatio-temporal characteristics of traffic flow (time dependence and space dependence) are not considered at the same time, and only the traffic flow data of a certain section are considered microscopically for prediction;
[0005] ② It does not take into account that the traffic flow is a gradually divergent flow process, that is, the traffic flow at the current location will gradually diverge from near to far as the time changes

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Traffic flow prediction method of divergence convolution recurrent neural network based on space-time diagram
  • Traffic flow prediction method of divergence convolution recurrent neural network based on space-time diagram
  • Traffic flow prediction method of divergence convolution recurrent neural network based on space-time diagram

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] This embodiment proposes a traffic flow prediction method based on the divergent convolutional neural network of the spatio-temporal graph. By analyzing and mining the collected traffic flow data, it is the main purpose of the traffic flow prediction to describe the urban road traffic conditions. In the data structure Among them, a graph is a very useful data structure, which is composed of a series of nodes and edge types, and has very obvious local connection properties. It is mainly used to represent the relationship between data. In the urban road traffic network, the intricate Urban roads, the intersection and separation of vehicles on the road, the entire traffic network is a complex system. Such as figure 1 As shown, the prediction method is as follows:

[0053] S1: Construct a directed weighted traffic network graph G according to the connection attributes of the traffic network, G=(v,ε,A), as the basic prediction unit. Where v is the predicted node in graph G...

Embodiment 2

[0100] The traffic speed data from Los Angeles for four consecutive months from March 1 to June 1, 2012 is used as a case test. Divide 207 prediction nodes, each node has a fixed latitude and longitude, and divide the time, every 5 minutes as a time interval, each road detection node has 288 records a day. A Traffic Flow Prediction Model Based on Set Prediction Nodes' Spatial Location and Temporal Relationship Component Graph Divergent Convolutional Recurrent Neural Networks. According to the change of time, the traffic flow can be divided into two kinds of traffic flow forecasts: steady trend and fluctuating trend. Among them, 14:00-16:00 belongs to the steady trend, and 8:00-10:00 belongs to the fluctuating trend. In order to analyze the correlation of traffic flow data in time and space, 70% of the data is used to train the model and estimate parameters, and 30% of the data is used as the verification data for prediction for comparative analysis. figure 2 and image 3 Tr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a traffic flow prediction method of a divergence convolution recurrent neural network based on a space-time diagram. The traffic flow prediction method is characterized in thata directed weighted graph of a road network is constructed based on spatial features of a traffic network, then a traffic flow prediction model of a graphic divergence convolution recurrent neural network is constructed by taking the directed weighted graph as a basic unit of prediction, deep learning is carried out by means of time-space characteristics of the traffic network, time-space prediction is carried out on the traffic flow of the traffic road network, a final traffic flow prediction model is constructed, and real-time prediction of the traffic flow is realized. The traffic flow prediction method has the advantages of precise prediction and high fitting degree.

Description

technical field [0001] The invention relates to the field of intelligent transportation, in particular to a traffic flow prediction method based on a spatio-temporal graph-based divergent convolutional cyclic neural network. Background technique [0002] Traffic flow prediction has important practical significance in the field of sustainable intelligent transportation. Traffic flow forecasting is one of the typical examples of this kind of continuous learning and improvement. Due to (1) the nonlinear time dynamics caused by the constantly changing road conditions; (2) the complex spatial dependence of the road network topology; With the rapid development of management technology, it is possible to process and apply massive traffic data in real time. Therefore, this chapter will combine one of the applications of big data—deep learning method, and propose a traffic flow prediction method based on graph divergent convolutional cyclic neural network , to predict the traffic f...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G08G1/01G06N3/04G06N3/08
CPCG08G1/0125G08G1/0145G06N3/08G06N3/045
Inventor 郑家佳吕建成谷振宇朱垚垚
Owner CHONGQING CITY MANAGEMENT COLLEGE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap