A New Method For Short-Term Traffic Flow Forecasting Based on Chaotic Time Series Analysis

摘要

On the basis of analysis of the time series of urban short-term traffic flow and the characteristic of chaos, this paper proposes a short-term traffic flow forecasting method through study on chaotic time series forecasting models. This method is based on adding-weight one-rank local-region (AWORLR) model. Weights are introduced and they make this method more adaptive and precise than whole-region methods. The forecasting result of actual short-time traffic flow is satisfying.