Robust Power Grid Optimization under Rising Ambient Temperature Scenarios

Abstract

Rising ambient temperatures driven by climate change introduce significant uncertainties and physical constraints into power grid operations and planning, simultaneously elevating electricity demand for cooling and degrading the performance of generation, transmission, and distribution assets. Higher temperatures increase conductor resistance and cause thermal expansion leading to sag in overhead lines, necessitating derating of static ratings or adoption of dynamic line ratings based on real-time weather. Thermal generation efficiency declines due to cooling limitations, solar photovoltaic output decreases with elevated panel temperatures, and overall system reserve margins tighten amid demand spikes. Traditional deterministic or stochastic optimization approaches often fail to guarantee feasibility or economic performance across a wide range of plausible temperature realizations. Robust optimization and distributionally robust optimization frameworks address these challenges by optimizing against worst-case scenarios within uncertainty sets or ambiguity sets derived from historical data and climate projections, ensuring solutions remain feasible and near-optimal under temperature dependent parameter variations. This research paper develops comprehensive robust power grid optimization models for unit commitment, optimal power flow, and integrated generation transmission planning that explicitly incorporate temperature impacts on line ampacity, generation limits, load profiles, and losses.