Online Robust Subspace Clustering With Application to Power Grid Monitoring

In this work, a robust subspace clustering algorithm is developed to exploit the inherent union-of-subspaces structure in data for reconstructing missing measurements and detecting anomalies. Our focus on processing an incessant stream of large-scale such as synchronized phasor power grid, which challenging due computational complexity, memory requirement, corrupt observations. order mitigate these issues, low-rank representation (LRR) model-based problem formulated that can handle sparse outliers data. Then, efficient online derived based stochastic approximation. The convergence property established. Strategies maintain representative yet compact dictionary capturing are also proposed. method tested both simulated real measurement unit (PMU) verify effectiveness shown significantly outperform existing algorithms simple