Integrating Deep Reinforcement Learning and Model Predictive Control for Microgrid Stability (DRIMPC-Stability)

This report presents a comprehensive analysis of a Hybrid Renewable Energy System (HRES) 
integrated into the IEEE 14-bus benchmark system. The study leverages the state-of-the-art Real
Time Digital Simulator (RTDS) and Hardware-in-the-Loop (HIL) technologies at RWTH-Aachen
to test and validate advanced control strategies for integrating renewable energy sources, 
including photovoltaic (PV) panels, wind turbines, and battery energy storage systems (BESS). 
The primary objective is to optimize energy management, enhance system stability, and address 
the operational challenges posed by renewable energy variability. 
The study begins by outlining the critical environmental challenges associated with greenhouse 
gas emissions and the urgent need to transition towards sustainable energy systems. Renewable 
energy sources and hybrid microgrids are identified as pivotal solutions to these challenges. The 
state-of-the-art section reviews existing advancements in renewable energy integration, control 
strategies, and the role of power electronics in improving system efficiency. It highlights gaps such 
as the underutilization of advanced control algorithms and the need for robust validation methods. 
In the experimental phase, the IEEE 14-bus system was modeled and simulated with the 
integration of renewable energy components. The HIL setup enabled real-time interaction 
between simulated and physical systems, allowing for high-fidelity testing of control strategies. 
Results demonstrated the feasibility and effectiveness of the proposed system in maintaining 
frequency and voltage stability while optimizing energy flow across the network. Key findings 
include improved system reliability and scalability for renewable energy integration. 
The analysis identified certain challenges, such as the limited adaptability of control algorithms 
and gaps in lab capabilities, particularly in real-time validation. Recommendations include 
enhancing the integration of dynamic control frameworks, expanding lab infrastructure, and 
refining the Holistic Test Description (HTD) to better align test objectives with experimental 
insights. These improvements will enable more robust and comprehensive testing for future 
projects. 
In conclusion, this project underscores the critical role of advanced simulation technologies and 
innovative control strategies in enabling the transition to sustainable energy systems. The 
outcomes provide a scalable model for integrating renewable energy into power grids, contributing 
to global efforts in achieving carbon neutrality. The ERIGrid 2.0 Lab Access program has proven 
instrumental in facilitating cutting-edge research and fostering collaboration between academic 
and industrial stakeholders. Future work should focus on addressing identified challenges and 
expanding the applicability of the developed methodologies to real-world scenarios.