Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Input-Output Specifications of Grid-Forming Functions and Data-Driven Verification Methods

Jennifer T. Bui, Dominic Groß

TL;DR

The paper addresses interoperability and performance verification for converter-interfaced generation (CIG) by developing decentralized, input-output based stability conditions that incorporate network circuit dynamics, and by formalizing disturbance-rejection performance specs. It introduces a technology-agnostic, small-signal model of a grid with network and device transfer functions, enabling device-level verification that holds across arbitrary topologies. A data-driven two-bus validation method recovers frequency response models from input-output data, enabling practical verification via Nyquist plots and EMT-based simulations. The approach highlights the roles of parameters such as line inductance, X/R ratio, coupling strength, and inner-loop bandwidth on GFM and GFL performance, and provides a scalable path toward standards-based interoperability for heterogeneous grid-forming resources. Potential extensions include phase-plot specifications, heterogeneous X/R networks, and incorporation of GFM PI controls.

Abstract

This work investigates interoperability and performance specifications for converter interfaced generation (CIG) that can be verified using only input-output data. First, we develop decentralized conditions on frequency stability that account for network circuit dynamics and can be verified using CIG terminal dynamics and a few key network parameters. Next, we formalize performance specifications that impose requirements on the CIG disturbance response. A simple data-driven validation method is presented that enables verification of the interoperability and performance specifications for CIG using input-output data from a two-node system. Data obtained from electromagnetic transient (EMT) simulations are used to illustrate the proposed approach and the impact of key parameters such as inner control loop gains, network coupling strength, and controller bandwidth limitations.

Input-Output Specifications of Grid-Forming Functions and Data-Driven Verification Methods

TL;DR

The paper addresses interoperability and performance verification for converter-interfaced generation (CIG) by developing decentralized, input-output based stability conditions that incorporate network circuit dynamics, and by formalizing disturbance-rejection performance specs. It introduces a technology-agnostic, small-signal model of a grid with network and device transfer functions, enabling device-level verification that holds across arbitrary topologies. A data-driven two-bus validation method recovers frequency response models from input-output data, enabling practical verification via Nyquist plots and EMT-based simulations. The approach highlights the roles of parameters such as line inductance, X/R ratio, coupling strength, and inner-loop bandwidth on GFM and GFL performance, and provides a scalable path toward standards-based interoperability for heterogeneous grid-forming resources. Potential extensions include phase-plot specifications, heterogeneous X/R networks, and incorporation of GFM PI controls.

Abstract

This work investigates interoperability and performance specifications for converter interfaced generation (CIG) that can be verified using only input-output data. First, we develop decentralized conditions on frequency stability that account for network circuit dynamics and can be verified using CIG terminal dynamics and a few key network parameters. Next, we formalize performance specifications that impose requirements on the CIG disturbance response. A simple data-driven validation method is presented that enables verification of the interoperability and performance specifications for CIG using input-output data from a two-node system. Data obtained from electromagnetic transient (EMT) simulations are used to illustrate the proposed approach and the impact of key parameters such as inner control loop gains, network coupling strength, and controller bandwidth limitations.
Paper Structure (27 sections, 2 theorems, 13 equations, 19 figures)

This paper contains 27 sections, 2 theorems, 13 equations, 19 figures.

Table of Contents

  1. Introduction
  2. Motivation and Problem Setup
  3. Power system model
  4. Network topology
  5. Device model
  6. Transmission line model
  7. Multi-converter / multi-machine network model
  8. Interoperability & Performance Specifications
  9. Interoperability of Frequency Dynamics
  10. Performance Specifications
  11. Example: GFM Droop control
  12. Example: SRF-PLL GFL control
  13. Example: GFM PI control
  14. Example: Interoperability with steady-state network model
  15. Example: Interoperability with network circuit dynamics
  16. ...and 12 more sections

Key Result

theorem 1

(Decentralized stability condition) Assume that all poles of $g_{\omega_n,p_n}(s)$ are in the open left half-plane for all $n \in \mathcal{N}$. The overall frequency dynamics shown in Fig. fig:smallsignalre are asymptotically stable if there exists $\alpha \in [0,\pi/2)$ such that holds for all $n \in \mathcal{N}$ and $\omega_p \in \mathbb{R} \cup \{\infty\}$.

Figures (19)

  • Figure 1: Power system consisting of generation units with synchronous generators and converters with different controls.
  • Figure 2: Abstracting the original network (top) through an analytical model (bottom) parameterized in key network parameters.
  • Figure 3: Verification of a decentralized stability condition for each individual device.
  • Figure 4: Converter-interfaced generation connected to a grid.
  • Figure 5: Small-signal frequency dynamics.
  • ...and 14 more figures

Theorems & Definitions (3)

  • theorem 1
  • proposition 1
  • proof