ARTIFICIAL INTELLIGENCE BASED HYBRID STRUCTURES FOR SHORT- TERM LOAD FORECASTING WITHOUT TEMPERATURE DATA

Abstract

Load forecasting is the first phase of electric power system planning for economic power generation-distribution, effective control and operation conditions of the system, and also energy pricing. In this study, short-term load forecasting, as the main tool for economic operation conditions, is realized. 24-hour-ahead load forecasting without temperature data for Turkey is aimed and structures with ANN, Wavelet Transform & ANN, Wavelet Transform & RBF Neural Network, and EMD & RBF Neural Network are proposed for forecasting process. Local holidays' load data is replaced with normal day's characteristic to remove the disturbing effects of those days. To have more accurate forecast, a regulation to load forecast is proposed. Unregulated and regulated forecast error percentages of all days except local holidays are calculated as average daily MAPE and maximum MAPE. All MAPE values are compared between the proposed structures.