Projects

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 studies the problem of localizing unknown disturbance sources in large linear dynamical systems using limited measurements. The work is motivated by oscillatory events, faults, and cyber-physical anomalies in power grids. ▣ Problem Disturbance inputs propagate through network dynamics Only partial measurements of the system state are available Need to infer where the excitation originated without full model knowledge ▣ Approach Apply subspace system identification to recover reduced-order dynamics Formulate a source localization estimator using Markov parameters Use structured propagation patterns to recover the most probable source ▣ Validation Demonstrated on synthetic linear networks and benchmark system models Accurate localization under noise, limited sensors, and model uncertainty Applicable to oscillation events and cyber-physical anomalies ▣ Publication IEEE CCTA 2023: https://ieeexplore.ieee.org/document/10252510

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

This project develops a Complex-LASSO framework for identifying the sources of forced oscillations (FOs) in wide-area power systems. ▣ Problem Forced oscillations degrade stability and can spread across large systems Source location is often unknown and difficult to estimate PMU measurements are noisy and distributed ▣ Approach Represent unknown FO inputs as sums of sinusoids in the frequency domain Estimate candidate frequencies via empirical spectrum analysis Formulate and solve a complex-valued ℓ₁-regularized least-squares problem Use coordinate descent for efficient optimization ▣ Validation IEEE 68-bus and WECC 179-bus systems Correctly identifies FO locations and amplitudes Scalable to multiple sources and noisy measurements ▣ Publication IEEE PES General Meeting (2022): https://ieeexplore.ieee.org/document/9916993

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)

This project applies adaptive sliding-mode control to virtual synchronous generators (VSGs) to improve microgrid frequency response and damping. ▣ Problem Converter-dominated microgrids lack physical inertia VSGs emulate synchronous behavior but need robust control Frequency excursions and oscillations persist under disturbances ▣ Approach Design an adaptive sliding-mode VSG controller Enhance frequency response and damping under varying conditions Evaluate robustness against load changes and disturbances ▣ Publication IEEE EPE (2018): https://ieeexplore.ieee.org/document/8559616

This project develops a probabilistic resilience assessment framework for distribution and microgrid systems under hurricane-induced disruptions. ▣ Problem Extreme weather events cause widespread outages Need quantitative metrics for expected resilience and restoration feasibility Load and PV generation uncertainty must be included ▣ Approach Define a time-dependent expected resilience metric Use Point Estimate Method (PEM) to propagate uncertainty Compare results against Monte Carlo simulation Apply graph-theoretic restoration feasibility analysis ▣ Validation IEEE 37-node distribution system Comparison of conventional vs. active distribution systems Strong agreement with Monte Carlo, with far lower compute cost ▣ Publication IEEE MELECON (2018): https://ieeexplore.ieee.org/abstract/document/8379107

This project develops a general resilience evaluation framework for radial distribution feeders exposed to extreme events. ▣ Problem Distribution networks experience structural vulnerabilities during storms and high-impact low-probability events. ▣ Approach Model resilience through topology, damage states, and restoration pathways Use graph metrics to quantify accessibility and load-serving potential ▣ Publication IEEE ICEE (2018): https://ieeexplore.ieee.org/document/8472496

This project applies NSGA-II multi-objective optimization to size battery storage and determine wind penetration levels for optimal microgrid performance. ▣ Problem Microgrid planning requires balancing cost, reliability, renewable penetration, and system constraints. ▣ Approach Formulate a multi-objective optimization problem Use NSGA-II to explore tradeoffs in storage and wind integration Analyze Pareto-front behavior across load and wind scenarios ▣ Publication IEEE ICEE (2017): https://ieeexplore.ieee.org/document/7985180