Collaborative Research: Auxiliary Signal-Based Fault Detection in Inverter-Dominated Power Systems

This NSF project aims to create new techniques for fault detection in inverter-dominated power systems. The most typical faults are unintentional short circuits caused by, for example, lightning and tree contact, and can cause substantial damage if not quickly dealt with. Detecting and extinguishing short-circuit faults are thus critical aspects of reliable power system operation. In power systems with synchronous machine-based generation, large, unbalanced fault currents provide clear information about the existence and location of faults. Inverters prevent such currents even during faults. As a result, conventional fault detection schemes can fail in grids that are rich in inverter-based resources like wind and solar generators. One way to make it easier to distinguish normal operation from a fault is for the inverter to add a perturbation, or auxiliary signal, when there is suspicion of a fault. This project will design new, auxiliary signal-based fault detection schemes for inverter-dominated power systems. The intellectual merits of the project include characterization of when and what auxiliary signals are necessary, how many inverters in a grid must inject them, and minimal infrastructure investments necessary to guarantee fault detection. The broader impacts of the project include enhanced reliability of modern power systems, which will further facilitate the integration of inverter-based resources like renewables and energy storage; and curriculum development at the graduate and undergraduate levels, and energy-oriented programming for K-12 students. This project will develop the use of auxiliary signals in inverter-dominated power systems. Typical choices of auxiliary signal include negative sequence current, as in IEEE Standard 2800, and harmonics. The auxiliary signals will be optimized so as to minimize disruption while guaranteeing detection of all possible faults. The existing theory will be extended in scenarios not covered by existing tools, for example, networks with multiple inverters and relays. The mathematical formalization of this problem will constitute a streamlined, optimization-based procedure for designing new detection schemes, which is relevant today as grids and grid codes continue to rapidly evolve. All new detection schemes will be validated in electromagnetic-transient simulation and in controller hardware-in-the-loop testing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.