Data-Driven Transmission Expansion Planning for Climate-Resilient Power Grids

Abstract

Transmission expansion planning traditionally optimizes the addition of new lines, reconductoring, or capacity upgrades to minimize long-term costs while meeting reliability and demand growth under deterministic or limited uncertainty assumptions. However, accelerating climate change introduces profound uncertainties through more frequent and intense extreme weather events particularly heatwaves, droughts, wildfires, and compound stresses that derate generation and transmission capacities, spike electricity demand, and heighten outage risks. Data driven approaches leverage high-resolution climate projections, historical weather datasets, real time sensor streams, and advanced analytics to embed these climate impacts directly into planning models. This research paper explores data-driven methodologies for TEP that enhance grid resilience, including distributionally robust optimization, stochastic programming with climate informed scenarios, integration of dynamic line ratings, and machine learning for uncertainty modeling and surrogate optimization. By co-optimizing generation, storage, and transmission under temperature-dependent parameters and extreme stress testing, these strategies reduce investment costs, improve renewable integration, lower operational expenses during heatwaves, and mitigate unserved energy. Case studies from the Western U.S. Interconnection demonstrate that cooperative, climate-adaptive planning can cut wholesale prices by over 60% and total system costs by 30%+ in future decades, even under widespread heat events.