Short-term Traffic Flow Forecasting Method Based on Interval Type-2 Fuzzy Neural Network

摘要

Interval type-2 fuzzy logic system cascaded with neural network, interval type-2 fuzzy neural system (IT2FNS), is proposed to handle complicated uncertainties in short-term traffic flow forecasting. A secondary membership function is obtained through fuzzy reasoning. The strong consistent estimates of the unknown parameters of the neural network structure are developed. The secondary membership function with upper and lower limit is utilized to create forecasting interval, which are suitable for handling complicated uncertainties. The efficiency and applicability of this forecasting technique is demonstrated by simulation results.