DEMAND FORECASTING BY USING GENERALIZED REGRESSION NEURAL NETWORK

Abstract

With the increases in competitive business environment, forecasting of demand is directly linked up with the
portfolio of the generating companies and considered as an important parameter of power system regulation. To take
decisions like bidding of the power blocks in a pool market, trade decisions, generation schedules, dispatch and unit
commitment operations require precise knowledge of the system demand. It is also essential to consider an accuracy of
load forecasting to maintain balance in power system regulation. In this paper we propose sensitivity analysis with an
application of adaptive supervised learning technique “Generalized Regression Neural Network” (GRNN) to know the
effect of the change in parameters in forecasting engine performance. Different inputs are changed uniformly and fair
comparison is carried out on the basis of standard error index Mean Absolute Percentage Error (MAPE).