Power Grid parameter estimation using Sparse Identification of Nonlinear Dynamics

Abstract

The recent discovery of nonlinear system identification via the Sparse Identification of Nonlinear Dynamics (SINDy) method has enjoyed a lot of success across many engineering applications. Due to innovations in sparse regression and compressed sensing, this technique enables tractable identification of both the structure and parameters of a nonlinear dynamical system from data. In the present work, we show the application of SINDy for estimating power-grid parameters. In particular, we demonstrate how SINDy can be used to extract the underlying swing equations from time-series data and thus estimate the critical power-system parameters like rotor inertia and damping coefficients with high degree of accuracy. We demonstrate the results on the Ring-Grid and the IEEE 39-Bus test system.