A new grey model for traffic flow mechanics. (February 2020)

• Record Type: Journal Article
• Authors: Xiao, Xinping; Duan, Huiming
• Is Part Of: Engineering applications of artificial intelligence. Volume 88(2020)
• Publication Date: 2020-02
• Subjects: Traffic flow mechanics -- Grey prediction model -- Short-term traffic flow forecasting -- Vehicle inflow rate -- Vehicle jam flow rate; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2019.103350
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Accurate and real-time short-term traffic flow prediction is the core technology of an intelligent transportation system. In this paper, the vehicle conservation principle of traffic flow mechanics is applied to study the differential equation of traffic flow is established by analysing traffic flow parameters. Using the principle of grey difference information, and a grey model of traffic flow in a road section is proposed. This model obtains traffic flow information about traffic flow inflow and congestion via matrix least squares technology and obtains the time response function and modelling steps of the model using a mathematical analysis method, which is applied to short-term traffic flow prediction. The results of three short-term traffic flow cases show that the simulation and prediction results of the new model are better than those of other grey models and two machine learning methods. Relevant information about the traffic flow parameters obtained by the new model is consistent with an actual situation of traffic flow. Highlights: The traffic flow differential equation is established. A new grey model of road traffic flow dynamics (TFDGM (1, 1)) is proposed. The model can effectively predict and quantitatively calculate the inflow rate