A Complex-LASSO Approach for Localizing Forced Oscillations in Power Systems

TL;DR

This paper introduces a complex-LASSO based method to localize multiple forced oscillation sources in power systems using noisy PMU data, leveraging sparsity and frequency domain analysis for improved accuracy.

Contribution

It develops a novel complex-LASSO approach for simultaneous localization and characterization of forced oscillations in power systems, addressing noise and multiple sources.

Findings

Efficient localization of FOs in simulated power systems.

Accurate estimation of oscillation frequencies, phases, and amplitudes.

Demonstrated effectiveness on IEEE and WECC test systems.

Abstract

We study the problem of localizing multiple sources of forced oscillations (FOs) and estimating their characteristics, such as frequency, phase, and amplitude, using noisy PMU measurements. For each source location, we model the input oscillation as a sum of unknown sinusoidal terms. This allows us to obtain a linear relationship between measurements and the inputs at the unknown sinusoids' frequencies in the frequency domain. We determine these frequencies by thresholding the empirical spectrum of the noisy measurements. Assuming sparsity in the number of FOs' locations and the number of sinusoids at each location, we cast the location recovery problem as an $\ell_1$-regularized least squares problem in the complex domain -- i.e., complex-LASSO (linear shrinkage and selection operator). We numerically solve this optimization problem using the complex-valued coordinate descent method, and show its efficiency on the IEEE 68-bus, 16 machine and WECC 179-bus, 29-machine systems.