A hybrid approach for short term electricity price and load forecasting

Abstract

In a deregulated power industry, accurate short term load forecasting (STLF) and price forecasting (STPF) is a key issue in daily power market. The load forecasting helps in unit commitment as well as in economic scheduling of the generators. The price forecasting helps an electric utility to make important decisions like generation of electric power, bidding for generation, price switching and infrastructure development. Price forecasting is very much useful for energy suppliers, ISOs and other participants in electric generation, transmission and distribution. This paper presents a hybrid approach for the STLF and STPF. The time series data pertaining to load / price is decomposed into various decomposition levels by the use of Wavelet Transform (WT) and each level obtained by this process is predicted using Artificial Neural Network (ANN). The performance of the proposed hybrid model is validated using New Delhi load data and Ontario electricity price data.