Short-term load forecasting of Toronto Canada by using different ANN algorithms

Abstract

In the field of power system, electrical load forecasting has got wide acceptance due to the determination of future trends of demand. Electrical load forecasting is used to predict the future power demand of the consumers which is estimated on the basis of the historical load data. The forecast model processes the exogenous relation of the provided data and consequently anticipates the future load demand. In this paper different ANN algorithms i.e Levenberg Marquardt back propagation, Bayesian Regularization & Scaled Conjugate Gradient algorithms has been applied for short-term load forecasting i.e. the one hour-ahead hourly forecast of the electricity load of Toronto, Canada using MATLAB R14a. The data used in the forecasting are hourly historical data of the temperature & the electricity demand of Toronto, Canada. The simulation results obtained is highly accurate and well forecasted.