Small-Signal Stability Condition of Inverter-Integrated Power Systems: Closed-Form Expression by Stationary Power Flow Variables

Taku Nishino, Yoshiyuki Onishi, Takayuki Ishizaki · 2025

Read Small-Signal Stability Condition of Inverter-Integrated Power Systems: Closed-Form Expression by Stationary Power Flow Variables by Taku Nishino, Yoshiyuki Onishi, Takayuki Ishizaki (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-20276.txt.

Introduction

Taku Nishino, Yoshiyuki Onishi, and Takayuki Ishizaki derive a stability test for power grids rich in inverters in a 2025 paper. As solar and battery units join the grid, grid staff need a clear rule for when small shakes will die out rather than grow.

Their rule uses only steady facts: part reactances, the network layout, and the resting power flow. It takes the form of matrix tests that each part can check in a shared way. A three bus test case shows the split in action: one kind of inverter load steadies the grid, while another kind upsets it.

Catalog record

CREATORTaku Nishino, Yoshiyuki Onishi, Takayuki Ishizaki
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-20276

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Small-Signal Stability Condition of Inverter-Integrated Power Systems: Closed-Form Expression by Stationary Power Flow Variables Taku Nishino, Yoshiyuki Onishi, Takayuki Ishizaki Abstract This paper shows that a necessary and sufficient condition for the small-signal stability of an inverter-integrated power system can be expressed in terms of semidefinite matrix inequalities determined only by the synchronous reactance of the components, the susceptance matrix of the transmission network, and the stationary values of the power flow distribution. To derive the stability condition, we consider a class of grid-forming inverters corresponding to a singular perturbation of the synchronous generator. The resulting matrix inequality condition, which has twice as many dimensions as the number of buses and is independent of the dynamics of the connected components, is expressed in terms of each component compensating in a decentralized manner for the loss of frequency synchronization caused by the reactive power consumption in the transmission network. A simple numerical example using a 3-bus power system model shows that a grid-forming inverter load improves power system synchronization, while a grid-following inverter load disrupts it. keywords Power systems, Small-signal stability, Equilibrium-independent passivity. 1 Introduction With the increase of DC power sources such as renewable energy and battery storage, there is a growing demand for stability analysis and stabilization control of power systems integrated with inverter resources ( Shah et al., 2015 ) . In particular, research on grid-forming inverters that simulate the physical properties of frequency synchronization of conventional synchronous generators has attracted much attention. The analysis of power systems with such inverters is based on different assumptions than those of conventional power systems with large inertia synchronous generators, and general-purpose numerical analyses such as eigenvalue calculations ( Sastry and Varaiya, 1980 ) and time-response calculations ( Stott, 1979 ) dominate. Against this background, this paper mathematically characterizes the small-signal stability of lossless power systems with synchronous generators and grid-forming inverters. The small-signal stability of power systems is equivalent to the local asymptotic stability of the equilibrium representing the power supply-demand balance. In particular, the following facts are proved in this paper. (i) A synchronous generator and a class of grid-forming inverters with identical synchronous reactances share the same decentralized stability condition. (ii) The stability condition can be expressed in a closed form using only the solution of an algebraic equation for the stationary power flow distribution. To clarify the contribution of this paper, we highlight its differences from existing small-signal stability analyses. The references ( Stegink et al., 2017 ; De Persis and Monshizadeh, 2017 ; Yang et al., 2019 ) analyze stability using passivity-based approaches for power systems consisting of one-axis synchronous generator models, inverter-based resources, and a mix of both, respectively. Similarly, this paper also adopts a passivity-based approach to prove the sufficiency of small-signal stability. However, none of the above references makes a clear distinction between the bus voltage and the internal voltage of the component connected to each bus. As a result, the mathematical model of the synchronous generator assumes dynamics that are partially inconsistent with the physics of the AC circuit. In particular, while the differential equation of the generator includes a certain reactance constant, the reactance in the equation for the connection between the generator and the bus, which should be assumed to have the same value, is implicitly assumed to be infinitesimal. Therefore, it is not possible to derive from the assumptions of these references the equivalence with respect to the synchronous reactance that this paper clarifies as fact (i). In addition, the derived stability condition cannot be expressed in a closed form of the stationary power flow distribution, as in fact (ii). It should be emphasized that this paper uses a standard synchronous generator model consistent with a proper AC circuit physics following ( Sauer et al., 2017 ) , and employs a more detailed two-axis model. Although omitted for space reasons, the same conclusions can be drawn using a more detailed Park generator model. This paper is organized as follows. Section 2 defines the power system model to be analized. Section 3 presents a necessary and sufficient condition for the small-signal stability and its proof. Section 4 shows a numerical example using a 3-bus power system model to demonstrate that a grid-forming inverter load improves power system synchronization, while a grid-following inverter load disrupts it. Section 5 concludes the paper. 2 Power System Model 2.1 Transmission Network Model Consider a transmission network with N buses. Without loss of generality, we assume that each bus is connected to a generator, load, or other device. Let b_{ij}>0 be the susceptance of the transmission line connecting Buses i and j . The susceptance matrix B\in\mathbb{R}^{N\times N} is given by B_{ij}=\left\{\begin{array}[]{cl}b_{ij},&\quad i\neq j\\ \displaystyle-\sum_{j=1}^{N}b_{ij},&\quad i=j.\end{array}\right. (1) This is negative semidefinite because -B is the weighted Laplacian of an undirected graph. Let \boldsymbol{I}_{i}\in\mathbb{C} and \boldsymbol{V}_{i}\in\mathbb{C} be the current and voltage phasors at Bus i , respectively. Assuming zero line conductances, i.e., lossless, the power balance at each bus is given as \left\{\>\begin{split}P_{i}&=\sum_{j=1}^{N}B_{ij}V_{i}V_{j}\sin(\theta_i -\theta_j)\\ Q_{i}&=\sum_{j=1}^{N}-B_{ij}V_{i}V_{j}\cos(\theta_i -\theta_j)\end{split}\right.\quad i\in\mathcal{N} (2) where \mathcal{N} is the label set of buses, V_{i} and \theta_{i} are the magnitude and phase of \boldsymbol{V}_{i} . Furthermore, P_{i} and Q_{i} are the active and reactive powers defined as the real and imaginary parts of \boldsymbol{V}_{i}\overline{\boldsymbol{I}_{i}} , respectively. The power balance equation in ( 2 ) can be simply written as 0=g(\theta,V,P,Q). (3) Since only the phase difference of voltage phasors matters, we introduce the following notion. Definition 1 Consider a stationary power flow distribution (\theta^{\star},V^{\star},P^{\star},Q^{\star}) such that 0=g(\theta^{\star},V^{\star},P^{\star},Q^{\star}). The set of equivalent stationary power flow distributions obtained by shifting voltage phases as (\theta_{\rm e}^{\star},V^{\star},P^{\star},Q^{\star})=\bigl\{(\theta^{\star}+c\mathds{1}_{N},V^{\star},P^{\star},Q^{\star}):c\in\mathbb{R}\bigr\} is called a stationary power flow distribution set . For simplicity, we use the notation of \varpi^{\star}:=(\theta^{\star},V^{\star},P^{\star},Q^{\star}),\quad\varpi_{\rm e}^{\star}:=(\theta_{\rm e}^{\star},V^{\star},P^{\star},Q^{\star}). In particular, we simply write f(\varpi_{\rm e}^{\star}) if a function f(\varpi^{\star}) is invariant for any \varpi^{\star} in \varpi_{\rm e}^{\star} . 2.2 Synchronous Generator Model In this paper, we consider the two-axis synchronous generator model ( Sauer et al., 2017 ) , which represents the electromechanical dynamics along two orthogonal axes called the d-axis and the q-axis. The dynamics of the two-axis generator model is given by \left\{\>\begin{split}\dot{\delta}_{i}&=\omega_{0}\omega_{i}\\ M_{i}\dot{\omega}_{i}&=-D_{i}\omega_{i}-P_{i}+P^{\star}_{\mathrm{m}i}\\ \tau_{\mathrm{d}i}\dot{E}_{\mathrm{q}i}&=-E_{\mathrm{q}i}-(X_{\mathrm{d}i}-X^{\prime}_{\mathrm{d}i})I_{\mathrm{d}i}+V^{\star}_{\mathrm{fd}i}\\ \tau_{\mathrm{q}i}\dot{E}_{\mathrm{d}i}&=-E_{\mathrm{d}i}+(X_{\mathrm{q}i}-X^{\prime}_{\mathrm{q}i})I_{\mathrm{q}i}\end{split}\right. (4a) where E_{\mathrm{q}i}\in\mathbb{R} is the q-axis internal voltage for the field winding, E_{\mathrm{d}i}\in\mathbb{R} is the d-axis internal voltage for the damper winding, \delta_{i}\in\mathbb{R} is the rotor angle relative to a reference frame rotating at the nominal angular frequency \omega_{0}\in\mathbb{R} , and \omega_{i}\in\mathbb{R} is the angular frequency deviation from \omega_{0} . For the positive constants, M_{i} represents the inertia constant, D_{i} the damping coefficient, \tau_{\mathrm{d}i} , \tau_{\mathrm{q}i} the time constants of the internal voltages, X_{\mathrm{d}i} , X_{\mathrm{q}i} the synchronous reactances X^{\prime}_{\mathrm{d}i} , X^{\prime}_{\mathrm{q}i} , the transient reactances, P^{\star}_{\mathrm{m}i} the mechanical power, and V_{\mathrm{fd}i}^{\star} the field voltage. The connection between the generator and bus is given by \textstyle I_{\mathrm{d}i}=\frac{1}{X^{\prime}_{\mathrm{d}i}}(E_{\mathrm{q}i}-V_{\mathrm{q}i}),\quad I_{\mathrm{q}i}=\frac{1}{X^{\prime}_{\mathrm{q}i}}(V_{\mathrm{d}i}-E_{\mathrm{d}i}) (4b) where V_{\mathrm{q}i} and V_{\mathrm{d}i} are defined as V_{\mathrm{q}i}=V_{i}\cos(\delta_i - \theta_i ),\quad V_{\mathrm{d}i}=V_{i}\sin(\delta_i - \theta_i ). Note that ( 4b ) can be expressed in terms of the active power P_{i} and reactive power Q_{i} supplied by the generator to the bus as \left\{\>\begin{split}P_{i}&=\textstyle\frac{E_{\mathrm{q}i}}{X^{\prime}_{\mathrm{d}i}}V_{\mathrm{d}i}-\frac{E_{\mathrm{d}i}}{X^{\prime}_{\mathrm{q}i}}V_{\mathrm{q}i}+\left(\frac{1}{X^{\prime}_{\mathrm{q}i}}-\frac{1}{X^{\prime}_{\mathrm{d}i}}\right)V_{\mathrm{d}i}V_{\mathrm{q}i}\\ Q_{i}&=\textstyle\frac{E_{\mathrm{q}i}}{X^{\prime}_{\mathrm{d}i}}V_{\mathrm{q}i}+\frac{E_{\mathrm{d}i}}{X^{\prime}_{\mathrm{q}i}}V_{\mathrm{d}i}-\left(\frac{V^{2}_{\mathrm{d}i}}{X^{\prime}_{\mathrm{q}i}}+\frac{V^{2}_{\mathrm{q}i}}{X^{\prime}_{\mathrm{d}i}}\right).\\ \end{split}\right. (4c) The synchronous and transient reactances generally satisfy X^{\prime}_{\mathrm{d}i}<X_{\mathrm{d}i},\quad X^{\prime}_{\mathrm{q}i}<X_{\mathrm{q}i}. (5) In the following, we represent the bus variables as v_{i}:=(\theta_{i},V_{i}),\quad w_{i}:=(P_{i},Q_{i}). Choosing them as the input and

Read the complete text