Utilisation of Deep Learning (DL) and Neural Networks (NN) Algorithms for Energy Power Generation: A Social Network and Bibliometric Analysis (2004-2022)

• Published in: International journal of energy economics and policy Vol. 14; no. 1; pp. 172 - 183
• Main Author: Zaidi, Abdelhamid
• Format: Journal Article
• Language: English
• Published: Mersin EconJournals 15.01.2024
• Subjects: Algorithms; Alternative energy; Bibliometrics; Collaboration; Deep learning; Energy economics; Machine learning; Neural networks
• ISSN: 2146-4553; 2146-4553
• DOI: 10.32479/ijeep.15161

The research landscape on the applications of advanced computational tools (ACTs) such as machine/deep learning and neural network algorithms for energy and power generation (EPG) was critically examined through publication trends and bibliometrics data analysis. The Elsevier Scopus database and the PRISMA methodology were employed to identify and screen the published documents, whereas the bibliometric analysis software VOSviewer was used to analyse the co-authorships, citations, and keyword occurrences. The results showed that 152 documents have been published on the topic comprising conference proceedings (58.6%) and articles (41.4%) between 2004 and 2022. Publication trends analysis revealed the number of publications increased from 1 to 31 or by 3,000% over the same period, which was ascribed to the growing scientific interest and research impact of the topic. Stakeholder analysis revealed the top authors/researchers are Anvari M, Ghaderi SF and Saberi M, whereas the most prolific affiliation and nations actively engaged in the topic are the North China Electric Power University, and China, respectively. Conversely, the top funding agency actively backing research on the topic is the National Natural Science Foundation of China (NSFC). Co-authorship analysis revealed high levels of collaboration between researching nations compared to authors and affiliations. Hotspot analysis revealed three major thematic focus areas namely; Energy Grid Forecasting, Power Generation Control, and Intelligent Energy Optimization. In conclusion, the study showed that the application of ACTs in EPG is an active, multidisciplinary, and impact area of research with potential for more impactful contributions to research and society at large.