Research on Electricity Price Forecasting Based on Chaos Theory

Abstract

Electricity price forecasting has attracted much interest in current literature. Many models and methods have been presented in the past, but most of them regarded the electricity price system as the linear system and using the linear theory to predict. In fact, electricity price system is a nonlinear system. According to its characteristics of complexity and non-linearity; this paper proposed a new electricity price forecasting model based on chaos theory. The proposed method makes use of chaos time series analysis to capture characteristics of complicated price behavior. First the chaotic characteristic of electricity price was verified based on phase-space reconstitution of time series by means of Lyapunov exponent. Then the improved add-weighted one-rank multi-steps prediction model is applied to a short-term electricity price forecasting. Comparing electricity price time series with historical date by chaos theory, the results indicated that it will be a bright merit to use chaos theory predicting electricity price of electricity industry.