Advanced voltage monitoring and control solutions
Quan Tran defended his PhD thesis at the Department of Electrical Engineering on December 3rd.
The rise of renewable energy and Distributed Energy Resources (DERs) marks a crucial shift toward a greener energy supply. However, this progress comes with challenges, including voltage violations and grid congestion, threatening the stability of distribution networks. In his PhD research Quan Tran delves into innovative strategies aimed at solving these challenges. The research proposes advanced voltage monitoring and control solutions, including neural network-based state estimation, adaptive DER control, and predictive modeling. Together, these innovations lay the foundation for more reliable, efficient, and resilient power grids.
Renewable energy sources and DERs are reshaping energy landscapes worldwide. Solar panels, wind farms, heat pumps, and electric vehicles contribute significantly to lowering carbon emissions. However, their variability and localized impacts often result in voltage imbalances and network congestion. As grids evolve into Active Distribution Networks (ADNs), these challenges underscore the need for robust monitoring and control mechanisms.
Building smarter grids
For Distribution System Operators (DSOs), real-time insights into grid performance are critical. This research highlights methods to enhance observability even with limited measurement data. Improved visibility allows DSOs to proactively address voltage and reliability issues, transforming traditional grids into dynamic, adaptive systems.
Neural network-based
A centerpiece of the thesis is a neural network-based Distribution System State Estimation (DSSE) model. By integrating grid topology into its algorithms, this model enhances voltage estimation in areas with sparse data. The approach promises significant improvements in grid stability while reducing the dependency on extensive sensor installations.
Manage voltage dynamics
Voltage control in active grids requires real-time responsiveness. introduces a Dynamic Mode Decomposition (DMD)-based predictive model to understand and manage voltage dynamics. This method equips operators with tools to anticipate changes, enabling smarter decision-making.
Accurate voltage predictions
During emergencies, quick, coordinated actions are crucial to prevent cascading failures. The thesis presents a Model Predictive Control (MPC) framework that utilizes Sparse Identification of Nonlinear Dynamics (SINDY) for accurate voltage predictions. This framework optimally coordinates DERs and On-Load Tap Changers (OLTCs), ensuring stability under pressure.
Real-world applications
While the findings demonstrate promise in simulations, bridging the gap to real-world applications remains a challenge. Future research should focus on realistic case studies and adherence to grid standards to make these innovations practical and impactful.
Title of PhD thesis: . Promotor: dr. Phuong Nguyen.