Content area
Full Text
Abstract
In this work, we present a three-stage stochastic optimization model for power grid resilience optimization against extreme weather events such as hurricanes and floods. The proposed framework integrates outputs from a predictive geoscience-based storm surge model with a direct current power flow model to identify substations susceptible to flooding and recommends hardening decisions. While doing so, it considers both the preparedness measures that can be taken before impending hurricanes and the recovery and social costs due to power outages during hurricane's aftermath. The model quantifies the contributions of hardening, preparedness, and recovery efforts in total resilience to a variety of storm surge scenarios and therefore provides explainable supports for optimal budget allocation for the three stages. We present a case study using a grid representative of Texas that demonstrates the efficacy of our model.
Keywords: Infrastructure Resilience, Extreme Weather, Storm Surge, Electric Grid, Hurricanes
(ProQuest: ... denotes formulae omited.)
1. Introduction
The Texas coast is frequently impacted by extreme flooding events. These events cause significant damage to various critical infrastructure networks such as transportation systems, healthcare services and the power grid. In 2017, Hurricane Harvey became not only the longest-lasting hurricane but also the costliest at $130B, part of which was due to power outages. Harvey damaged more than 90 substations, 800 transmission assets, 6000 distribution poles, and 800 miles of power lines, with a peak power generation loss of 11GW, affecting more than 2 million people. It took 2 weeks and 12,000 crew members to restore power [1]. In order to make the power grid resilient to such extreme events, researchers have proposed a variety of decision-making models for different problems in the mitigation, preparedness, response, and recovery phases of the power grid resilience management cycle [3], [2], [10] [9] [5]. Most of these models define a problem in a specific phase and assume a certain upper bound on some available resource that it has to optimally allocate within the power grid network. However, determining the upper bound on the resource that should be made available in the different phases of resilience management is one of the most important questions to address in order to leverage the aforementioned models for resilience decision-making. To the best of our knowledge, there are no studies...