Paper Info
Title
An inertia grey discrete model and its application in short-term traffic flow prediction and state determination
Abstract
A traffic flow system is a complex dynamic system. Traffic flows data are the product of the velocity and density, and its data have dynamic and fluctuation characteristics. Therefore, three new inertia grey discrete models (IDGMs) were proposed and used to estimate short-term traffic flow based on traffic flow data mechanics and characteristics and traffic-state characteristics. The modelling process of the traditional grey DGM using the least square method may lead to a large parameter estimation deviation and a low model precision. The new model uses the mechanical characteristics of the data and applies the evolutionary process of the mechanical decomposition of the data to the modelling process. It has a more reasonable modelling process and a more stable structure and solves the shortcomings of the traditional grey DGM parameter estimation. Moreover, it uses matrix analysis to study the important characteristics of the IDGM, and it simplifies the forms of the parameter model and structural model. Then, the traffic flow of the Whitemud Drive City Expressway in Canada is analysed empirically, and the effect of the new model and the judgment of three-phase traffic flow state are analysed.
Year
DOI
Venue
2020
10.1007/s00521-019-04364-w
NEURAL COMPUTING & APPLICATIONS
Keywords
DocType
Volume
Grey prediction model,Short-term traffic flow forecasting,Inertia model,Force resolution,Traffic flow state
Journal
32.0
Issue
ISSN
Citations 
SP12
0941-0643
1
PageRank 
References 
Authors
0.36
0
3
Name
Order
Citations
PageRank
Huiming Duan133.11
Xinping Xiao251.47
Qinzi Xiao310.36