Power Systems Engineering Research Center

Project Summaries

Power Systems

Sparse Sensing Methods for Model-Free Sensitivity Estimation and Topology Change Detection using Synchro-Phasor Measurements (S-59)

Summary Among the computational methods developed to treat large statistical time series, techniques for 'sparse inverse covariance estimation' have seen tremendous progress in the literature of the last decade. Phasor Measurement Unit (PMU) data share many of the features of general statistical time series, with the added characteristic that phasor voltage magnitude and phase angle are highly influenced by the underlying power flow equations for the network being measured. This project will exploit sparse inverse covariance methods for 'model-free' estimation of the power flow Jacobian and related sensitivity matrices, in near real time. Additionally, by leveraging the inherent variability in the voltage magnitude and phase angle time series provided by PMUs, the project will also develop similarly fast time-scale (i.e., second or sub-second update rates, much faster than state estimator update cycles) filters for detection and identification of topology changes, exploiting the fact that topology changes will appear as low rank changes in the subspace spanned by the phasor measurement time series.
Academic Team Members Project Leader: Alejandro D. Domínguez-García (University of Illinois at Urbana-Champaign, aledan@illinois.edu)
Team members: Pete Sauer (University of Illinois at Urbana-Champaign, psauer@illinois.edu);
Chris DeMarco (University of Wisconsin-Madison, demarco@engr.wisc.edu);
Steve Wright (University of Wisconsin-Madison, swright@cs.wisc.edu)
Industry Team Members Mirrasoul Mousavi (ABB); Prashant Kansal (AEP); Jim Kleitsch (ATC); Jim Gronquist (BPA); Evangelos Farantatos (EPRI); Alan Engelmann (Exelon-ComEd); Slava Maslennikov (ISO-NE); George Stefopoulos (NYPA); Angel A. Aquino-Lugo (PowerWorld); Mahendra Patel (PJM); Jay Caspary (SPP)