Application of Machine Learning in Energy Systems – a Comparative Analysis of Three Case Studies

The exponential growth in the number of papers published annually in the field of machine learning
applications in energy systems presents a challenge to researchers seeking to conduct comprehensive and
effective literature reviews. To address this issue, we took a systematic literature review approach with three
distinct smaller case studies focusing on the application of machine learning in energy systems, namely:
1. Machine learning in drilling
2. Machine learning for rooftop solar energy potential quantification, and
3. Machine learning in district heating and cooling in the context of seasonal thermal energy storages.
In each case, we employed a systematic literature review methodology. For topic one, we utilized an existing
comprehensive review to generate further insights and information. For topics two and three, we used
predefined search criteria to conduct relevant publications in a systematic and reproducible manner. We
investigate the state of the art of the use of machine learning in these distinct areas of inquiry, thereby
facilitating the identification of research gaps. Ultimately, we compare approaches and models utilized in each
field, identified common best practices, and propose methods to address potential challenges.