A Model-Agnostic Method for PMU Data Recovery Using Optimal Singular Value Thresholding

This paper presents a fast model-agnosticmethod for recovering noisy Phasor Measurement Unit (PMU) data streams with missing entries. The measurements are first transformed into Page matrix, and the original signals reconstructed using low-rank matrix estimation based on optimal singular value thresholding. Two variations of recovery algorithm shown- a) an offline block-processing method imputing past measurements, b) online predicting future measurements. Information within PMU channel (temporal correlation) as well from different channels in network (spatial utilized to recover degraded data. proposed is needs no explicit knowledge underlying system model or measurement noise distribution. performance algorithms illustrated simulated IEEE 39-bus test real anonymized U.S. electric utility. Extensive numeric tests show that can be accurately recovered presence additive noise, consecutive drop, simultaneous erasures across multiple channels.