Demand Response Management of Grid-Connected Nano-Grid Using Data Driven Algorithms

The use of renewable energy resources (RERs) has drawn more attention in recent decades and extensive research has been conducted to determine how best to integrate them into the electrical infrastructure to maximize their benefits. Managing demand response (DR), which entails adjusting the burden to match the availability of RERs, is one of the most difficult aspects of integrating RERs. This paper proposes a DR optimization technique based on real-time prices (RTP) for wind and solar-powered DC residences. The proposed strategy utilizes machine learning (ML) techniques including Decision Tree (DT) Regressors, Random Forest (RF), and Linear Regression (LR) on a multi-source, renewable energy sources inclusive dataset to maximize profit via load shifting. A novel approach to tune the hyperparameters of ML algorithms with an Antlion optimizer is used in this research to enhance their efficacy. Our study compares the proposed RTP-based dynamic pricing technique to conventional fixed pricing strategies to determine its effectiveness. The simulation results indicate that the proposed method outperforms conventional fixed pricing strategies.