Machine-Learning Applications in Energy Efficiency: A Bibliometric Approach and Research Agenda

Abstract

The high demand for energy resources due to the increasing number of electronic devices has prompted the constant search for different or alternative energy sources to reduce energy consumption, aiming to meet the high demand for energy without exceeding the consumption of natural sources. In this context, the objective of this study was to examine research trends in the machine-learning-based design of electrical and electronic devices. The methodological approach was based on the analysis of 152 academic documents on this topic selected from Scopus and Web of Science in accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement. Quantity, quality, and structural indicators were calculated to contextualize its thematic evolution. The results showed a growing interest in the subject since 2019, mainly in the United States and China, which stand out as world powers in the information and communication technology industry. Moreover, most studies focused on developing devices for controlling, monitoring and reducing energy consumption, mainly in 5G and thermal comfort devices, primarily using deep-learning techniques.