Resilience Constrained Day-Ahead Unit Commitment Under Extreme Weather Events

Abstract

Over the last years, extreme weather events have caused extensive damages in power systems, leaving millions of customers without electricity and therefore highlighting the necessity to enhance power system resilience. This paper proposes a resilience constrained day-ahead unit commitment framework for increasing resiliency of a power system exposed to an extreme weather event. The weather-dependent failure probabilities of the transmission lines are taken into account in order to decide the scheduling of generators that minimizes load shedding in the most efficient way, while respecting operating limits of the system. The problem is formulated as a tri-level optimization problem that is transformed to a bi-level problem using duality theory and linearization techniques. The problem is solved as a two-stage robust optimization problem using a Column & Constraint Generation based decomposition algorithm. The master problem provides the unit commitment and the subproblem identifies the worst damage scenario due to weather event. A Sequential Monte Carlo simulation of a modified IEEE Reliability Test System and IEEE 118-bus System is applied to illustrate and validate the effectiveness of the proposed framework.