Digital twins for energy management: bridging virtual intelligence and physical systems

The implementation of Digital Twin (DT) systems is increasingly being embraced across various fields, especially in energy management and promoting sustainable development. It is an emerging technology across industries, which helps with improving decision-making, decreasing carbon footprint, and improving energy efficiency. This research focuses on the examination of Digital Twins (DTs) in the energy management field across various industries. The objective of this research is to consolidate and analyze existing academic and peer-reviewed literature on the topic through bibliometric methods and structured analytical synthesis. The findings indicate that the earliest studies in the final reviewed corpus on the integrated use of DT, AI, and IoT for energy-related applications were published in 2019, and that the manufacturing industry was among the earliest sectors to adopt DT-based approaches with a primary focus on improving energy efficiency. The building industry, however, has become the most active field of study in the application of DTs for advancing sustainability. Additionally, China stands out as one of the leading contributors to this growing field. In terms of methodological techniques, supervised learning emerges as the primary approach in both Machine Learning (ML) and Deep Learning (DL). Furthermore, Deep Reinforcement Learning (DRL) algorithms are preferred over traditional Reinforcement Learning (RL) algorithms, highlighting a shift towards more adaptable and scalable solutions in intricate settings. These observations offer important insights into prevailing trends and propose potential future research paths at the intersection of digital technology and sustainable development.