Advanced Short-Term Load Forecasting in Power Systems Using Machine Learning Technologies

The current research is a holistic examination of Short-term-load-forecasting (STLF) with the aid of machine learning methods, with emphasis on the Panama case study. The data used cover several data sources, such as historical electricity load data, weather factors, and calendar factors. Machine learning models, such as multiple linear regression and neural networks, are utilized for accurate forecasting of electricity load demand. The research also compares the model predictions with official predictions in an effort to determine their effectiveness. The results confirm that machine learning algorithms in STLF work well and point to the relevance of including contextual variables for achieving higher forecasting accuracy. Additionally, the paper explores the method's limitations and recommends future research avenues to make STLF models more reliable and scalable. In general, this study helps to pave the way for developing STLF methodologies with the potential benefits of using machine learning methods in load forecasting.