A stochastic optimization framework to planing for geographically correlated failures in coupled natural gas and electric power systems

Abstract

We develop a transmission planning framework for coupled natural gas and electric power systems facing geographically correlated failures. The framework uses a stochastic optimization method incorporating uncertainty in the locations of the geographically correlated failures. We compare the proposed planning framework with the traditional N-k method which plans for geographically uncorrelated failures in terms of improving resilience effectively against geographically correlated failures. The proposed planning framework is illustrated using a small test system, but is scalable to larger system sizes and portable to other coupled-infrastructure contexts. We illustrate how planning for geographically correlated failures can enhance system-wide robustness by reinforcing the network connectivity between supply and demand locations instead of addressing local vulnerability at specific locations when planning for geographically uncorrelated failures. Planning for geographically correlated failures improves system resilience to both geographically correlated and uncorrelated failures significantly, but the converse is not true in our case study.