Application of Machine Learning Techniques in Energy Power Production: A Publication Trend and Bibliometrics Analysis (2012-2023)

Abstract

An analysis of publishing trends and bbibliographic analysis data was conducted to critically analyse the research landscape regarding the utilisation of machine learning techniques in the field of energy and power production (EPP). The Elsevier Scopus database and the PRISMA methodology were utilised to locate and evaluate the published papers. The bibliometric analysis software package RStudio (Biblioshiny) was employed to examine the most significant sources, authors, and institutions with the highest productivity. The findings indicated that a total of 653 documents were published on the subject, consisting of conference proceedings (41.8%) and articles (50.8%), spanning the years 2012 to 2023. An review of publishing patterns indicated a significant increase in the number of publications, rising from 3 to 190, representing a growth of 5,000% over the same period. This surge can be attributed to the expanding scientific interest and the influence of research on the subject. The stakeholder analysis identified Boumaiza A, Sanfilippo A, and Wang X as the leading authors/researchers in the field. Additionally, it revealed that China, India, and the United States are the most actively involved nations in this area. In contrast, the primary financial organisation that actively supports study on the subject is the National Natural Science Foundation of China (NSFC). Overall, the study demonstrated that the utilisation of machine learning in evidence-based policy and practice (EPP) is a dynamic and interdisciplinary field of research that has the potential to make significant contributions to both research and society.