Electric Load Forecasting Using Machine Learning and Traditional Models: A Review

Abstract

Electric load forecasting (ELF) is an essential procedure in the electricity industry's planning, with a significant impact on electric capacity scheduling and power systems management. As a result, it has garnered growing attention from the academic community. Load forecasting has become an essential component of electricity utility firms. In order to ensure uninterrupted and reliable power supply to consumers, dlecision-makers in the utility sector need to accurately predict the future electricity demand, minimizing any margin of error. Therefore, the precision of electric load forecasting is crucial for scheduling energy generation capacity and managing power systems. This paper examines the techniques and a model used for predicting electricity load, and provides an overview of the latest advancements in electric load forecasting technologies. It focuses on recent studies that have explored the combination of multiple machine learning algorithms to create hybrid models. A total of 44 academic articles were utilized to compare various projects based on specific criteria, including the time frame, project size, and the methods and models implemented.