Research progress and prospects of machine learning applications in renewable energy: a comprehensive bibliometric-based review

The stability of power system operations is being challenged by the rapid development of renewable energy. A viable solution is to achieve accurate renewable energy forecasting. In this regard, machine learning has attracted widespread attention from researchers. However, there remains a lack of comprehensive and in-depth research summarizing the progress of machine learning in renewable energy (ML & RE). This study conducted a bibliometric analysis of 1804 publications (2012–2023) in the ML & RE field to obtain an overview of the research progress and status, as well as proposing future research directions. The main results indicate that (1) the ML & RE research field encompasses a variety of disciplines and is experiencing a significant increase in publications. (2) China, the United States, India, Saudi Arabia and their universities, have a broad foundation of cooperation and mature research experience in this field. However, at the micro level, scholars are in a dispersed state and have weak cooperation with each other. (3) Developing models and algorithms to enhance the accuracy of renewable energy forecasting remains a key focus of current research. (4) Forecasting under multiple renewable energy hybrid scenarios, algorithm updates and optimizations based on new data, and consideration of the impact of green electricity trading and carbon trading policies, etc., will be potential directions for future research. This study aids researchers in comprehending the progress and prospects of the ML & RE research field.