Identifying Unique Power System Signatures for Determining Vulnerability of Critical Power System Assets

摘要

In this paper, the finer granularity of phasor measurement unit (PMU) data is exploited to develop a data-driven approach for accurate health assessment of large power transformers (LPTs). The research demonstrates how variations in signal-to-noise ratios (SNRs) of PMU measurements can be used as a reliable metric for health assessment. However, a single PMU device may be affected by multiple equipment located in its neighborhood. To address the challenge of identifying the equipment that is primarily responsible for the degradation in quality of the PMU measurements, an intelligent sensor selection scheme is proposed, which ensures that every critical equipment is associated with a unique identifying signature. The proposed algorithm is based on a real LPT failure event that occurred in the US Southwest. The inferences drawn from the proposed PMU-based health monitoring scheme can be easily supplemented with other LPT sensors to facilitate proactive intervention before the point-of-no-return is reached.