Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Momentum-Accelerated Online Feedback Optimization for Power System Flexibility

Florian Klein-Helmkamp, Matthis Berger, Irina Zettl, Andreas Ulbig

TL;DR

This paper addresses real-time coordination of distributed flexibility in power systems by enhancing Online Feedback Optimization with a momentum-augmented projected gradient approach. The method leverages both current and past gradient information to accelerate convergence while enforcing feasibility via a quadratic projection step. Two case studies—online congestion management on a MV distribution feeder and multi-layer flexibility dispatch across grid layers—demonstrate faster convergence and robust constraint satisfaction, even under disturbances. The results show that proper tuning of the momentum and gain parameters yields significant improvements in responsiveness, suggesting momentum-augmented OFO as a viable tool for scalable real-time flexibility control in large-scale power networks.

Abstract

Flexibility is increasingly gaining importance in modern power system operation. This paper presents a controller framework based on Online Feedback Optimization for real-time coordination of power system flexibility. The proposed approach introduces a momentum-augmented projection-step to accelerate convergence and improve dynamic performance. We derive the controller formulation, and evaluate its performance and stability in two representative case studies. The first examines online congestion management in distribution feeders, and the second addresses multi-layer flexibility dispatch across system interfaces. Numerical results demonstrate that the momentum-based controller achieves faster convergence and maintains constraint satisfaction, highlighting its potential for real-time flexibility control in large-scale power systems.

Momentum-Accelerated Online Feedback Optimization for Power System Flexibility

TL;DR

This paper addresses real-time coordination of distributed flexibility in power systems by enhancing Online Feedback Optimization with a momentum-augmented projected gradient approach. The method leverages both current and past gradient information to accelerate convergence while enforcing feasibility via a quadratic projection step. Two case studies—online congestion management on a MV distribution feeder and multi-layer flexibility dispatch across grid layers—demonstrate faster convergence and robust constraint satisfaction, even under disturbances. The results show that proper tuning of the momentum and gain parameters yields significant improvements in responsiveness, suggesting momentum-augmented OFO as a viable tool for scalable real-time flexibility control in large-scale power networks.

Abstract

Flexibility is increasingly gaining importance in modern power system operation. This paper presents a controller framework based on Online Feedback Optimization for real-time coordination of power system flexibility. The proposed approach introduces a momentum-augmented projection-step to accelerate convergence and improve dynamic performance. We derive the controller formulation, and evaluate its performance and stability in two representative case studies. The first examines online congestion management in distribution feeders, and the second addresses multi-layer flexibility dispatch across system interfaces. Numerical results demonstrate that the momentum-based controller achieves faster convergence and maintains constraint satisfaction, highlighting its potential for real-time flexibility control in large-scale power systems.

Paper Structure

This paper contains 14 sections, 9 equations, 10 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Main contribution
  4. Controller Framework
  5. General Optimization Problem
  6. Online Feedback Optimization
  7. Momentum-Accelerated Projected Gradient Descent
  8. Case Study I: Congestion Management
  9. Problem Setup
  10. Results and Analysis
  11. Case Study II: Flexibility Dispatch
  12. Problem Setup
  13. Results and Analysis
  14. Conclusion

Figures (10)

  • Figure 1: Closed-loop implementation of projected gradient descent.
  • Figure 2: Used configuration of CIGRE MV benchmark system cigre_2014. Open switches are indicated by boxes.
  • Figure 3: Bus voltages during online congestion management with generation disconnection event at $k=24$.
  • Figure 4: In-feed and consumption of active and reactive power of flexible loads and generators for $\beta=0.9$.
  • Figure 5: Hierarchical structure of interacting OFO controllers.
  • ...and 5 more figures