Studying load forecasting techniques in power system and their applications

Abstract

The basis of this project is to evaluate the effectiveness of the load forecasting methods and to determine their efficiency in providing accurate forecasts. The first phase of the project was focused on the theory behind the different load forecasting methods that are existing in the market. In the next phase, short-term load forecasting models were programmed. In this research, 8 models were constructed based on 7 different techniques. The techniques are Auto-Regressive Integrated Moving Average (ARIMA), Artificial Neural Network (ANN), Support Vector Machine (SVM), Recurrent Neural Network (RNN), Kalman Filtering, and lastly Gaussian Process.