Enhancing Lossy Compression of PMU Measurements by Data Conditioning

Synchrophasors generate large volumes of data that greatly benefit from compression for its efficient storage and transfer. In this paper, we investigate the effect of the quality of phasor measurement data on lossy compression, and propose effective data conditioning approaches to improve the compression performance. Specifically, three conditioning methods are employed: a Hamming window-based low-pass filtering scheme to clean measurements corrupted by noise, an algorithm to identify and eliminate measurements with significant amount of noise, and a simple scaling scheme to normalize the magnitudes of different types of measurements. Numerical results obtained on PMU data showcase the capability of the proposed approaches in improving compression performance.