Event Identification

This project introduces a semi-supervised learning framework for classifying disturbance events using high-dimensional PMU data when labels are sparse. Real power-system PMU archives contain thousands of unlabeled snapshots and only a limited number of trusted labeled events. ▣ Problem PMU archives contain very limited labeled data, making supervised training unreliable Grid conditions shift across time, seasons, and topology changes Physics-based features alone cannot fully capture dynamic variability ▣ Approach Combine spectral/modal signatures with learned temporal embeddings Apply self-training, pseudo-label refinement, and consistency constraints Leverage small labeled sets to guide structure in large unlabeled PMU corpora ▣ Validation Strong generalization under low-label regimes Robust identification across multiple utilities and operating conditions Outperforms traditional supervised PMU classifiers in label-scarce settings ▣ Publication arXiv preprint (2024): https://arxiv.org/abs/2309.10095

This project develops a real-time event identification framework using modal and spectral features extracted from wide-area PMU streams. ▣ Problem PMUs produce extremely high-dimensional, high-rate data Operators need rapid recognition of line trips, generator outages, and oscillations SCADA-level alarms do not capture modal characteristics ▣ Approach Extract modal features (frequency, damping, mode shapes) Build a high-dimensional spectral–temporal representation Apply feature selection (LASSO, MI ranking) Train logistic regression and SVM classifiers ▣ Validation Texas 2000-bus synthetic grid Real utility dataset with ≈500 PMUs Strong accuracy and robustness across conditions ▣ Publication IEEE Transactions on Power Systems (2022): https://ieeexplore.ieee.org/document/9911774

A poster contribution demonstrating the use of modal analysis and spectral PMU features for disturbance detection and classification. ▣ Problem Wide-area events leave modal signatures that are not captured by traditional SCADA indicators. ▣ Approach Extract modal dynamics from PMU windows Build feature embeddings for event classification Evaluate temporal stability of modal signatures ▣ Publication IEEE PES General Meeting Poster (2021)