A Novel Approach for Electricity Demand Forecasting using ANN

Abstract

Demand forecasting plays an important role in electricity management system. The forecasting results can be used to determine the vulnerability of the system. There are various techniques that can forecast the demand. These techniques include parametric model, SVM, neural network techniques and genetic algorithms etc. In our approach, we are using ANN for forecasting. The ANN model allows more flexible relationship between demand and weather variables as compared to multiple regression model. Our approach performs well as compared to regression model. In this paper, we have discussed the results of artificial neural network technique for monthly demand forecasting for the Amloh city.