A review of short-term power load and renewable energy generation forecasting based on machine learning
Download as PDF
DOI: 10.25236/icceme.2024.020
Corresponding Author
Bowei Deng
Abstract
Smart grid is an important development direction in today's energy field. With the rapid development of smart grid technology, the prediction of short-term power load and renewable energy generation has become a key link of smart grid. The prediction of short-term power load, wind energy and solar energy is crucial to the efficient and sustainable operation of power grid. Machine learning, as a powerful data analysis tool, has shown great potential in the field of power system forecasting in recent years. In this context, this paper presents a bibliometric analysis of the literature in this field over the last 15 years and describes some of the machine learning methods used in the literature, which were specifically developed for power load forecasting and power forecasting for wind turbines and solar panels. In addition, this paper compares different machine learning methods and discusses their similarities, with the aim of providing researchers with the choice of machine learning methods. This paper also discusses some of the data sets used in the literature that adapt to their proposed machine learning methods, so that researchers can better choose the appropriate data sets in future work. Finally, the challenges and prospects in the field of short-term power load and renewable energy forecasting are presented.
Keywords
Machine learning, renewable energy, power load, forecasting