Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Model-Free Power System Stability Enhancement with Dissipativity-Based Neural Control

Yifei Wang, Han Wang, Kehao Zhuang, Keith Moffat, Florian Dörfler

TL;DR

The paper tackles transient stability in power systems with converter-interfaced generation by proposing a model-free, dissipativity-based control framework for grid-connected VSGs. It learns dissipativity-encoding matrices V, Q, S, and R from input–state data using matrix neural networks and synthesizes a stabilizing feedback π(x) = −R^{-1}(x)S^T(x)x without requiring a full dynamical model. Stability is guaranteed through dissipativity conditions, while cost shaping enables alignment with user-defined performance goals, yielding an effective infinite-horizon stabilizing control under the learned model. Numerical tests on a single-converter infinite bus and a four-VSG Kundur two-area system demonstrate enlarged regions of attraction and robust post-fault synchronization, highlighting the method's practical potential for modern low-inertia grids.

Abstract

The integration of converter-interfaced generation introduces new transient stability challenges to modern power systems. Classical Lyapunov- and scalable passivity-based approaches typically rely on restrictive assumptions, and finding storage functions for large grids is generally considered intractable. Furthermore, most methods require an accurate grid dynamics model. To address these challenges, we propose a model-free, nonlinear, and dissipativity-based controller which, when applied to grid-connected virtual synchronous generators (VSGs), enhances power system transient stability. Using input-state data, we train neural networks to learn dissipativity-characterizing matrices that yield stabilizing controllers. Furthermore, we incorporate cost function shaping to improve the performance with respect to the user-specified objectives. Numerical results on a modified, all-VSG Kundur two-area power system validate the effectiveness of the proposed approach.

Model-Free Power System Stability Enhancement with Dissipativity-Based Neural Control

TL;DR

The paper tackles transient stability in power systems with converter-interfaced generation by proposing a model-free, dissipativity-based control framework for grid-connected VSGs. It learns dissipativity-encoding matrices V, Q, S, and R from input–state data using matrix neural networks and synthesizes a stabilizing feedback π(x) = −R^{-1}(x)S^T(x)x without requiring a full dynamical model. Stability is guaranteed through dissipativity conditions, while cost shaping enables alignment with user-defined performance goals, yielding an effective infinite-horizon stabilizing control under the learned model. Numerical tests on a single-converter infinite bus and a four-VSG Kundur two-area system demonstrate enlarged regions of attraction and robust post-fault synchronization, highlighting the method's practical potential for modern low-inertia grids.

Abstract

The integration of converter-interfaced generation introduces new transient stability challenges to modern power systems. Classical Lyapunov- and scalable passivity-based approaches typically rely on restrictive assumptions, and finding storage functions for large grids is generally considered intractable. Furthermore, most methods require an accurate grid dynamics model. To address these challenges, we propose a model-free, nonlinear, and dissipativity-based controller which, when applied to grid-connected virtual synchronous generators (VSGs), enhances power system transient stability. Using input-state data, we train neural networks to learn dissipativity-characterizing matrices that yield stabilizing controllers. Furthermore, we incorporate cost function shaping to improve the performance with respect to the user-specified objectives. Numerical results on a modified, all-VSG Kundur two-area power system validate the effectiveness of the proposed approach.
Paper Structure (15 sections, 18 equations, 6 figures, 1 algorithm)

This paper contains 15 sections, 18 equations, 6 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Power System Transient Stability
  4. Lyapunov Function and Dissipativity
  5. Methodology
  6. Dissipativity-Based Stabilizing Control
  7. The Matrix Neural Networks
  8. Loss Function Design
  9. Learning Dissipativity with Neural Networks
  10. Simulation Validation
  11. Single Converter Infinite Bus System
  12. Kundur Two-Area System
  13. Conclusions
  14. Proof of Theorem 2
  15. Proof of Theorem 3

Figures (6)

  • Figure 1: The architecture of three types of matrix NNs. For simplicity, we only draw one MLP schematic. Different matrix NNs do not share the same MLP.
  • Figure 2: An overview of our algorithm, which generate a stabilizing control after training NNs that can characterize dissipativity.
  • Figure 3: Schematic of the SCIB system.
  • Figure 4: The effect of proposed control in the SCIB system.
  • Figure 5: The modified Kundur two-area system with VSGs.
  • ...and 1 more figures