ARTIFICIAL NEURAL NETWORK BASED SHORT TERM LOAD FORECASTING FOR RESTRUCTURED POWER SYSTEM

Abstract

Electric Load forecasting and Electric Price forecasting are an important component in the economic and secure operation of the restructured power system energy management. A number of electric power companies are equipped to make forecasts with the aid of traditional statistical methods. This dissertation presents the use of an artificial neural network to a day ahead load forecasting, half hour ahead load forecasting and half hour ahead price forecasting applications for the restructured power system. By using historical weather, load consumption, price and calendar data, a multi-layer feed forward (FF) neural network trained with Back propagation (BP) algorithm are developed for these applications. The developed algorithm for a day ahead forecasting has been tested with IIT Roorkee campus data. The half hour ahead load and price forecasting algorithm has been tested with Australian market data. The results of ANN forecasting models are compared with the conventional Multiple Regression (MR) forecasting model. The aim of this dissertation is to examine the forecasting ability of a neural network in a situation where the electric load was subject to considerable seasonal variations over the year. The variations are affected by energy demand related to the climatic conditions.