Management and real-time monitoring of interconnected energy hubs using digital twin: Machine learning based approach

Abstract

Getting equipped by highly new smart technologies, Energy Hubs (EHs) and Smart Grids (SGs) are gaining interest these days. Energy management will advance over time as a result of the interaction impact among power and natural gas grids, and the use of smart technology for communications. The present study proposes a novel approach entitled Smart EH (SEH) for modeling multi-carrier energy systems in SG environments. Furthermore, this paper determines the optimum management and sizing of combined heat and power, auxiliary boiler, absorption chiller, as well as transformer unit as the essential components of an SEH. It is difficult to address the requirements of SGs with most conventional load scheduling algorithms because they lack robustness and performance in complex environments. An evaluation of the benefits and costs of optimizing such parameters was carried out in this paper and the Reinforcement Learning (RL) algorithm is applied to solve the optimization problem. An individual user in a dynamic electrical market was examined as an SEH in support of the suggested approach. According to simulation outcomes, the suggested method is effective regarding time efficiencies and load variations.