Online learning for robust voltage control under uncertain grid topology
2306.16674
0
0
Abstract
Voltage control generally requires accurate information about the grid's topology in order to guarantee network stability. However, accurate topology identification is challenging for existing methods, especially as the grid is subject to increasingly frequent reconfiguration due to the adoption of renewable energy. In this work, we combine a nested convex body chasing algorithm with a robust predictive controller to achieve provably finite-time convergence to safe voltage limits in the online setting where there is uncertainty in both the network topology as well as load and generation variations. In an online fashion, our algorithm narrows down the set of possible grid models that are consistent with observations and adjusts reactive power generation accordingly to keep voltages within desired safety limits. Our approach can also incorporate existing partial knowledge of the network to improve voltage control performance. We demonstrate the effectiveness of our approach in a case study on a Southern California Edison 56-bus distribution system. Our experiments show that in practical settings, the controller is indeed able to narrow the set of consistent topologies quickly enough to make control decisions that ensure stability in both linearized and realistic non-linear models of the distribution grid.
Create account to get full access
Overview
- This research paper explores how to maintain stable voltage levels in electrical power distribution grids, even when the grid topology (how components are connected) is uncertain or changing.
- The authors propose an "online learning" approach that allows the control system to adapt and adjust in real-time as the grid topology changes.
- Key techniques used include convex body chasing, an optimization method, and reinforcement learning to train the control system.
Plain English Explanation
Electrical grids are complex systems with many components like power lines, transformers, and loads (things consuming electricity). The way these components are connected, known as the grid topology, can change over time as the grid expands or components fail.
Maintaining stable voltage levels across the grid is critical, as too high or too low voltage can damage equipment. However, adapting the voltage control system to an uncertain, changing topology is challenging.
The researchers developed a control system that can learn and adapt in real-time as the grid topology changes. It uses an optimization technique called "convex body chasing" to track the optimal voltage control settings. The system also employs reinforcement learning, which is like trial-and-error learning, to improve its performance over time.
This allows the voltage control system to be robust to grid topology changes, ensuring consistent power quality for consumers even as the underlying grid infrastructure evolves. The online learning approach is key, as it enables the control system to react quickly to changes without human intervention.
Technical Explanation
The paper proposes an online learning framework for robust voltage control in distribution grids with uncertain topology. The core technical components are:
-
A convex body chasing optimization algorithm that tracks the optimal voltage control settings as the grid topology changes. This involves solving a sequence of convex optimization problems.
-
A reinforcement learning agent that learns a control policy to adjust voltage set points. The agent receives feedback on the grid's voltage profile and adjusts its actions accordingly to improve performance.
-
A simulation environment that models a distribution grid with uncertain, time-varying topology. This allows the researchers to train and evaluate the performance of the online learning control system.
Through extensive simulations, the authors demonstrate that their approach can maintain stable voltages even as the grid topology changes significantly over time. The online learning enables the control system to continuously adapt without requiring manual reconfiguration.
Critical Analysis
The paper presents a promising approach to a practically important problem. The online learning framework seems well-designed to handle the challenges of uncertain, changing grid topologies. The use of convex optimization and reinforcement learning are well-justified technical choices.
However, the paper does not address how the proposed system would be implemented in practice. Issues like communication latency, hardware constraints, and integration with existing grid control infrastructure are not discussed. Real-world deployment would likely require further engineering considerations.
Additionally, the simulation environment, while useful for evaluating the concept, may not fully capture the complexities of actual distribution grids. Testing on real-world grid data or hardware-in-the-loop setups could provide additional insights.
Future research could also explore the scalability of the approach as grid sizes increase, as well as its robustness to other types of uncertainty beyond topology changes, such as renewable energy intermittency or unexpected load changes.
Conclusion
This research proposes an innovative online learning framework for maintaining robust voltage control in evolving power distribution grids. By combining convex optimization and reinforcement learning, the system can adapt in real-time to changes in grid topology, a critical capability as grids become more dynamic and complex.
While further work is needed to address practical deployment considerations, the core technical ideas presented in this paper represent an important step towards making power grids more resilient and adaptable. As the energy landscape continues to transform, control systems that can learn and adjust autonomously will be invaluable for ensuring reliable, high-quality electricity service.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
Distributionally Robust Policy and Lyapunov-Certificate Learning
Kehan Long, Jorge Cortes, Nikolay Atanasov
0
0
This article presents novel methods for synthesizing distributionally robust stabilizing neural controllers and certificates for control systems under model uncertainty. A key challenge in designing controllers with stability guarantees for uncertain systems is the accurate determination of and adaptation to shifts in model parametric uncertainty during online deployment. We tackle this with a novel distributionally robust formulation of the Lyapunov derivative chance constraint ensuring a monotonic decrease of the Lyapunov certificate. To avoid the computational complexity involved in dealing with the space of probability measures, we identify a sufficient condition in the form of deterministic convex constraints that ensures the Lyapunov derivative constraint is satisfied. We integrate this condition into a loss function for training a neural network-based controller and show that, for the resulting closed-loop system, the global asymptotic stability of its equilibrium can be certified with high confidence, even with Out-of-Distribution (OoD) model uncertainties. To demonstrate the efficacy and efficiency of the proposed methodology, we compare it with an uncertainty-agnostic baseline approach and several reinforcement learning approaches in two control problems in simulation.
4/8/2024
🌿
A Predictive Operation Controller for an Electro-Thermal Microgrid Utilizing Variable Flow Temperatures
Max Rose, Christian A. Hans, Johannes Schiffer
0
0
We propose an optimal operation control strategy for an electro-thermal microgrid. Compared to existing work, our approach increases flexibility by operating the thermal network with variable flow temperatures and in that way explicitly exploits its inherent storage capacities. To this end, the microgrid is represented by a multi-layer network composed of an electrical and a thermal layer. We show that the system behavior can be represented by a discrete-time state model derived from DC power flow approximations and 1d incompressible Euler equations. Both layers are interconnected via heat pumps. By combining this model with desired operating objectives and constraints, we obtain a constrained convex optimization problem. This is used to derive a model predictive control scheme for the optimal operation of electro-thermal microgrids. The performance of the proposed operation control algorithm is demonstrated in a numerical case study.
4/3/2024
🧠
Graph neural networks for power grid operational risk assessment under evolving grid topology
Yadong Zhang, Pranav M Karve, Sankaran Mahadevan
0
0
This article investigates the ability of graph neural networks (GNNs) to identify risky conditions in a power grid over the subsequent few hours, without explicit, high-resolution information regarding future generator on/off status (grid topology) or power dispatch decisions. The GNNs are trained using supervised learning, to predict the power grid's aggregated bus-level (either zonal or system-level) or individual branch-level state under different power supply and demand conditions. The variability of the stochastic grid variables (wind/solar generation and load demand), and their statistical correlations, are rigorously considered while generating the inputs for the training data. The outputs in the training data, obtained by solving numerous mixed-integer linear programming (MILP) optimal power flow problems, correspond to system-level, zonal and transmission line-level quantities of interest (QoIs). The QoIs predicted by the GNNs are used to conduct hours-ahead, sampling-based reliability and risk assessment w.r.t. zonal and system-level (load shedding) as well as branch-level (overloading) failure events. The proposed methodology is demonstrated for three synthetic grids with sizes ranging from 118 to 2848 buses. Our results demonstrate that GNNs are capable of providing fast and accurate prediction of QoIs and can be good proxies for computationally expensive MILP algorithms. The excellent accuracy of GNN-based reliability and risk assessment suggests that GNN models can substantially improve situational awareness by quickly providing rigorous reliability and risk estimates.
5/14/2024
🏅
End-to-End Reinforcement Learning of Curative Curtailment with Partial Measurement Availability
Hinrikus Wolf, Luis Bottcher, Sarra Bouchkati, Philipp Lutat, Jens Breitung, Bastian Jung, Tina Mollemann, Viktor Todosijevi'c, Jan Schiefelbein-Lach, Oliver Pohl, Andreas Ulbig, Martin Grohe
0
0
In the course of the energy transition, the expansion of generation and consumption will change, and many of these technologies, such as PV systems, electric cars and heat pumps, will influence the power flow, especially in the distribution grids. Scalable methods that can make decisions for each grid connection are needed to enable congestion-free grid operation in the distribution grids. This paper presents a novel end-to-end approach to resolving congestion in distribution grids with deep reinforcement learning. Our architecture learns to curtail power and set appropriate reactive power to determine a non-congested and, thus, feasible grid state. State-of-the-art methods such as the optimal power flow (OPF) demand high computational costs and detailed measurements of every bus in a grid. In contrast, the presented method enables decisions under sparse information with just some buses observable in the grid. Distribution grids are generally not yet fully digitized and observable, so this method can be used for decision-making on the majority of low-voltage grids. On a real low-voltage grid the approach resolves 100% of violations in the voltage band and 98.8% of asset overloads. The results show that decisions can also be made on real grids that guarantee sufficient quality for congestion-free grid operation.
6/21/2024