Research on Traffic Flow Prediction based on Chaotic Time Series

In order to dramatically improve the precision of traffic flow estimates, this study proposes a prediction technique based on chaotic time series. Here is a breakdown of the exact research process. First, the chaos principle of traffic flow and two indexes-delay time and dimension selection-that aff...

Full description

Saved in:
Bibliographic Details
Published inIAENG international journal of applied mathematics Vol. 53; no. 3; pp. 236 - 240
Main Authors Yang, Xiaobo, Liu, Lianggui
Format Journal Article
LanguageEnglish
Published Hong Kong International Association of Engineers 01.09.2023
Subjects
Accuracy
Algorithms
Chaos theory
Delay time
Expected values
Neural networks
Support vector machines
Time series
Traffic flow
Online AccessGet full text

Cover

More Information
Summary:In order to dramatically improve the precision of traffic flow estimates, this study proposes a prediction technique based on chaotic time series. Here is a breakdown of the exact research process. First, the chaos principle of traffic flow and two indexes-delay time and dimension selection-that affect system reconstruction are looked at. Then, in order to obtain the maximum amount of time that can be forecasted, the chaotic traffic flow is predicted using an improved local approach and the Lyapunov index. Finally, a comparison is made between the traffic flow predictions made using the traditional local technique and the modified local method. According to the outcomes of the predictions, chaotic time series can be utilized to forecast traffic flow, and the prediction error is lower than that of both the widely used neural network prediction method and the least squares support vector machine prediction method, proving the effectiveness of the method proposed in this study.
ISSN:1992-9978
1992-9986