Energy Management

Microgrids, including terrestrial systems, shipboard power systems, port microgrids, and community-scale grids, have undergone rapid transformation in recent years due to increasing electrification and the integration of distributed energy resources (DERs). These developments, coupled with advanced energy management (EM) systems, enable more flexible, efficient, and intelligent operation of modern power networks. Advanced control and optimization architectures play a critical role in enhancing microgrid performance across multiple dimensions, including fuel efficiency, operational cost, reliability, resilience, and environmental sustainability. However, EM problems in such systems are often inherently nonconvex, large-scale, and subject to uncertainty, making real-time implementation particularly challenging. In our lab, we develop learning-enabled optimization frameworks that combine model-based control with data-driven techniques to address these challenges. Our approaches are designed to efficiently solve complex, nonconvex EM problems in real time, enabling adaptive and reliable operation under dynamic conditions. By leveraging advances in machine learning and computational optimization, we aim to support the next generation of resilient, intelligent, and sustainable energy systems.