SHORT TERM LOAD FORECASTING METHODS, A COMPARATIVE STUDY

Electrical & Electronics Engineering

National Conference on "Recent Advances in Materials Science & its Applications" & "Recent Trends in Electrical, Electronics, Computing and Information Technology"

Load forecasting is a very important function in electrical power systems. Accurate load forecasting is the need for economic cost saving. Short-Term Load Forecasting (STLF) is one of the forecasting methods that have a time frame of a few hours to about a day. STLF is required for adequate scheduling and operation of power systems. For the past several decades, there have been various methods used in load forecasting. Being a highly required activity both from a power system and economic point of view, extensive research is being made in the area of load forecasting. Some of the methods proposed earlier, and still being used are Box-Jenkins models, Auto Regressive Integrated Moving Average (ARIMA) models, Kalman filtering models, and the spectral expansion techniques-based models. Recent techniques are based on Artificial Intelligence methods like artificial neural networks (ANN), fuzzy logic, expert systems, support vector methods (SVM). Each technique has its own advantages and shortcomings. This paper does a comparative study of the methods with their positives and limitations.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.