Type-2 Fuzzy Forecasting of Main Road Traffic Flow

Abstract

In the process of development of modern intelligent transportation system, how to effectively deal with the nonlinear, time-varying and sudden traffic flow data and accurately predict the traffic flow is one of the important technologies to achieve traffic control and guidance. Now most of the methods are through the shallow model in learning, it is to easy to fall into the local extreme phenomenon which can not be more complex mathematical operations. In order to overcome the limitation of the traditional forecasting technology on the random variation of the massive traffic flow data, this paper proposes a traffic flow forecasting method based on the interval type-2 fuzzy sets. Firstly, stratified sampling and K-means clustering are combined to collect the data of the roads in all directions of the main road. The central limit theorem is used to convert the scattered data points into the interval form first, and the fuzzy set and type-2 fuzzy set, and through fuzzy to get the forecast result finally. The experimental results show that the result obtained in this paper is a range of forecasting region, which is difficult to be achieved by traditional methods and has high accuracy and reliability.