| dc.description.abstract | Due to the adverse effects of geomagnetic disturbances (GMDs) and geomagnetically induced currents (GICs) on the electric power grid there is increased interest in effective means of monitoring and visualizing such effects. As mitigation plans are now mandated for grid operators and research is being pursued regarding system operations and control during a GMD, the ability to know and understand the current state of the system in real time, especially with respect to GICs, is increasingly needed. To address these challenges, a GIC state estimation method is developed, improved upon, and tested. Additional applications to enhance situational awareness and inform decision making during a GMD are covered, including bad data detection, improvement upon the traditional state estimation methodology, and integration with existing power system analysis and interactive software. A GIC estimation method is proposed which leverages neutral GIC measurements and available electric field data to estimate the underlying electric field. Distinct from the traditional ac state estimation problem, this dc method leverages the linear relationship between the measurements and newly-defined states for a fast and efficient solution. The performance of the estimator is tested under different scenarios, such as varying measurement error, measurement availability, and granularity of electric field states. The estimator is shown to be an effective means of providing system-wide understanding throughout these scenarios. The concepts of observability and redundancy are defined for this new estimation method, laying a foundation for an extension to bad data detection. The least absolute value objective is invoked and identification thresholds designed such that outlier data in the measurement set can be detected, identified, and suppressed. Bad measurements are able to be reliably identified, even in the presence of multiple bad data. The traditional state estimation methods are extended to include GIC modeling, measurements, and states. It is shown to improve the performance in terms of average error and convergence time during a GMD. Other applications and extensions are explored, including non-Gaussian noise considerations, comparison of available real-time electric field data, and real-time visualization of GICs from real data. | en |
