Artificial intelligence-powered energy community management for developing renewable energy systems in smart homes

Abstract

The increase in demand due to scientific innovation and economic expansion raises challenges linked to environmental impact and sustainable market development. This study analyzes the role played by Peer-to-Peer (P2P) energy trading in system operation. Several case studies with four distinct prosumer models, each fitted with diverse Energy Storage Systems (ESS), photovoltaic (PV) systems, and responsive demand technologies such as Electric Vehicles (EV), are examined. An economic analysis is conducted using the developed interface. An energy management system establishes a dynamic market framework, enabling energy trading via P2P market operators and distribution system operators. The implemented trading platform is chosen depending on pricing variations across different time intervals for home models. The P2P energy trading system employs mathematical modeling via deep learning to achieve an optimal solution that minimizes costs. The study uses other domestic models and optimization techniques to explore the amount of energy traded across households and timeframes. Additionally, it investigates the accomplishment of two-way energy transfer in electric cars, the importance of energy storage and photovoltaic systems, and various housing scenarios concerning peer-to-peer energy trading. Additionally, the system's advantages are demonstrated to participants beforehand by validating and presenting data through an input and output system that accepts these instances.