Short-Term Load Forecasting using Artificial Neural Network in Indonesia

Short-term Load Forecast (STLF) is a load forecasting that very important to study because it determines the operating pattern of electrical system. Forecasting errors, both positive and negative, result in considerable losses costs increase ultimately lead waste. STLF research Indonesia, especially State Electricity Company (PLN Sulselrabar), has yet be widely used. Methods mainly used are manual conventional methods they considered adequate. In addition, Indonesia's geographical conditions extensive diverse, electricity system complex. As result, factors affecting each country's demand different, so unique needed. Artificial Neural Network (ANN) one Intelligent (AI) for can model complex non-linear relationships from networks. This paper aims build an suitable using several ANN models tested. Based on models, test results obtained best Model-6 with architecture (9-20-1). hidden layer, 20 neurons sigmoid logistic activation function (binary sigmoid), linear function. performance values mean squared error (MSE), absolute (MAE), percentage (MAPE) 430.48 MW2, 15.07 MW, 2.81%, respectively.