Applications of Digital Twins for Demand Side Recommendation Scheme with Consumer Comfort Constraints.

The next evolution of traditional energy systems towards smart grid will require end-consumers to actively participate and make informed decisions regarding their energy usage. Industry 4.0 facilitates such progress by allowing more advanced analytics and creating means for end-consumers and distributed grid assets to be modelled as their Digital twins (DT) equivalents, paving the way for asset-level analytics. Note-worthily, consumers’ comfort is crucial towards promotion of easy adoption of such models from consumers’ perspectives. This study presents the application of hybrid DT and multiagent reinforcement learning models for real-time estimation of end-consumers future energy behaviors while generating actionable recommendation feedback for improving their energy efficiency and enhancing end-user comfort.