A Machine Learning-Aided DERMS to Manage Large Numbers of Small-Scale Distributed Energy Resources

The rapid rise in the penetration of distributed energy resources (DERs) has introduced significant challenges in modern power systems. The integration of numerous small-scale DERs further complicates the landscape, leading to increased complexity and making these challenges more difficult to address. This paper seeks to address the limitations of existing approaches by introducing a three-level Distributed Energy Resources Management System (DERMS). The proposed DERMS aggregates numerous small-scale DERs at the bottom level and manages them effectively alongside individual DERs at the top level, offering a more coordinated and efficient solution. Additionally, the Random Forest Regression (RFR) model has been selected as the machine learning model to simplify the optimal power flow (OPF) problem using a Sequential Adjustment Method (SAM), significantly enhancing the computational efficiency of the process. The effectiveness and robustness of the proposed DERMS framework have been demonstrated through several complex case studies simulated on the modified IEEE 123-bus test system, highlighting its potential to improve the reliability and stability of power systems in the face of increasing DER penetration. Furthermore, this approach provides a scalable solution that can adapt to the evolving demands of modern power grids, ensuring sustainable and reliable energy management in the future.