Source-linked AI summary
From Rigid to Adiabatic: Canonical Regularization of AC Networks via Action-Angle Variables
Feng Ji, Lu Gao, Lihui Yang
TL;DR
The paper addresses the limits of rigid-network, timescale-separated AC models as converter controls interact with transmission-line magnetic dynamics. It models rotating magnetic fields with action-angle coordinates to obtain a canonical port-Hamiltonian network formulation, then studies a classical two-node system through Bloch-sphere energy landscapes. The resulting framework conserves total action, links reactive-power imbalance to angle evolution, and exposes saddle-point instability under voltage control while connecting P-δ and Q-V stability analyses.
Problem
Rigid algebraic power-flow models assume fast electromagnetic transients, but converter penetration creates comparable network and control timescales that challenge this approximation.
Method
The paper uses action-angle canonical coordinates for rotating magnetic fields, regularizes AC-network constraints into port-Hamiltonian dynamics, and analyzes a classical two-node system with Bloch-sphere coordinates.
Results
The formulation reveals conserved total nodal action, reactive-power-driven angle evolution, and saddle-point instability associated with reactive-power voltage control.
Takeaways & Limitations
The regularized framework connects electromagnetic-energy dynamics with conventional power-flow and equal-area analyses, unifying P-δ angle and Q-V voltage stability views within the studied system.
Abstract
from arXiv · showhide
Traditional power system analysis relies on timescale separation and the rigid-network assumption, freezing electromagnetic transients into algebraic power-flow equations via Steinmetz's phasor theory. As grid-forming converter penetration increases, magnetic energy dynamics on transmission lines interact with converter control loops on comparable timescales, challenging this rigid-network assumption. Returning to Faraday's law of electromagnetic induction driven by rotating magnetic fields, this paper models the transmission lines' rotating magnetic fields in action-angle canonical coordinates, regularizes the rigid algebraic constraints of power-flow equations into canonical equations on adiabatic symplectic manifolds, and establishes a port-Hamiltonian standard form for AC power grids. Based on the minimal-counterexample principle and using the equal-area criterion's classical two-machine system, this paper reveals a latitudinal instability channel via Bloch-sphere coordinates: Q-V control releases voltage-amplitude freedom, shifting the stability boundary from the equatorial UEP (unstable equilibrium point) to a saddle point, thereby unifying the analytical frameworks of P-delta angle stability and Q-V voltage stability in power system analysis.
I. INTRODUCTION
The paper replaces rigid algebraic AC-network constraints with an action-angle, port-Hamiltonian formulation that restores electromagnetic energy dynamics. In a classical two-node system, it links reactive-power imbalance to angle evolution and uses Bloch-sphere energy landscapes to connect angle and voltage stability.
- Motivation: Traditional analysis compresses electromagnetic transients into algebraic power-flow constraints under timescale separation, an assumption challenged by converter-grid timescale coupling.High grid-forming-converter penetration can make line magnetic-energy dynamics and inverter-control dynamics comparable, while quasi-steady-state equivalents may miss transient behavior.
- Canonical formulation: The paper identifies nodal action and angle as canonical coordinates and establishes a port-Hamiltonian standard form for AC networks.This canonical regularization restores network constraints as dynamic equations driven by Hamiltonian energy gradients.
- Conservation laws: The total nodal action is conserved during adiabatic processes, equivalently preserving the network’s mean squared nodal-voltage magnitude.The paper interprets this invariant as an energy-level coordination objective for grid-forming converter clusters rather than only local voltage regulation.
- Stability landscape: Bloch-sphere coordinates expose energy-landscape equilibria and a saddle-point instability associated with reactive-power voltage control.The classical equatorial stable and unstable equilibrium points agree with the equal-area criterion, while the paper uses the landscape to relate P-δ and Q-V stability.
- Stability landscape: In steady state, the regularized differential model reduces to the classical power-angle and reactive-power equations.This connects the dynamic formulation to conventional power-flow analysis while retaining electromagnetic-energy evolution during transients.
- Conservation laws: Reactive-power imbalance directly drives relative-angle evolution until reactive-power balance is restored at a new steady operating point.The framework identifies an indirect transient chain in which active-power change induces reactive-power imbalance and subsequent angle drift, which algebraic power-flow equations do not capture.
V MODIFICATION OF THE ENERGY LANDSCAPE BY ACTIVE‑POWER CONTROL
The active-power P–δ control modifies the energy landscape and shifts the equilibrium points while preserving equatorial motion when voltage magnitudes remain constrained. Its steady-state landscape yields a critical-energy boundary consistent with the equal-area criterion.
- P–δ control: The active-power control energy function includes a correction term proportional to Prefδωref.
- Equilibrium shifts: For Pref=2 MW, the SEP shifts from 0° to 41.29° and the UEP from 180° to 138.71°.Their energy difference is 3.67 kJ, while two symmetric high-latitude saddle points also appear.
- Equatorial constraint: With constant voltage magnitudes, the system remains confined to the equator, so the high-latitude saddle points do not cause instability.
- Stability test: A 3.58 kJ disturbance remains stable because it exceeds the saddle-point threshold but stays below the equatorial UEP critical energy of 3.67 kJ.The time-domain simulation therefore shows no first-swing instability.
- Q–V modification: Introducing Q–V control releases latitudinal motion, shifts the SEP away from the equator, and makes lower-energy saddle points the stability boundary.In the reported case, the initial 3.58 kJ exceeds saddle thresholds of 2.10 kJ and 2.52 kJ but remains below the 3.84 kJ UEP threshold.
- Network extension: The action-angle model extends from the two-node system to n-node, m-branch networks while conserving total action in the undamped case.Total-action conservation corresponds to invariant mean squared nodal-voltage magnitude.
APPENDIX A
Appendix A derives the voltage relations from complex flux-linkage dynamics in rotating coordinates. Taking real parts produces the first equation of (A.2).
- Complex representation: The complex flux linkage is represented by magnitude and angle, with Ψ=Ψ_x+jΨ_y and θ=angle(Ψ).
- Real-part derivation: Taking the real part of the flux-linkage relation yields u=2Lω(u_x cosθ+u_y sinθ).
- Result: The appendix concludes that this expression proves the first equation of (A.2).
A.2 Prove the second equation of (A.2):
Appendix A derives the second equation of (A.2) by using the logarithmic derivative of the complex flux linkage and extracting its imaginary part.
- Logarithmic derivative: For Ψ=Ψe^{jθ}, the logarithmic derivative separates relative magnitude change from angular change.
- Imaginary-part derivation: Taking the imaginary part produces the voltage relation involving ux, uy, θ, and ω.
- Result: The appendix concludes that the derived relation proves the second equation of (A.2).
APPENDIX B
Appendix B shows that the two-node dynamic model reduces to conventional power-flow equations when dynamic processes are neglected.
- Equivalence target: The appendix poses equivalence between the rotating-coordinate equation and conventional power flow for a two-node, single-branch system.
- Quasi-static reduction: Neglecting dynamic processes reduces the flux-linkage dynamics to an algebraic relation.
- Admittance form: Equation (B.4) yields the conventional nodal admittance equation.
- Power-flow equations: Combining the nodal admittance equation with nodal injection powers gives the conventional power-flow algebraic equations.