Predictive Mitigation of Short Term Voltage Instability Using a Faster Than Real-Time Digital Replica

Abstract

Predictive mitigation of undesired events has long been seen as a supportive complement to corrective mitigation that could relax the stringent requirements on the corrective actions and increase reliability of the overall system. This article describes one such predictive measure, i.e. the use of faster than real-time simulation in detecting faults and predicting the dynamic behavior for the resilient operation of future smart grid systems. A predictive mitigation strategy is proposed for a fault induced dynamic voltage recovery (FIDVR) event. These events, although rare, are typically addressed with under voltage load shedding schemes (UVLS) which leave significant portion of load under-supplied. We show that, by using the digital faster than real-time replica, the minimal level of UVLS can be determined on-the fly as the event develops while ensuring only the minimal amount of load shed.