Impact of Heatwaves on Power System Reliability and Mitigation Through Data-Driven Optimization

Abstract

Heatwaves, intensified by climate change, pose one of the most severe threats to power system reliability by simultaneously escalating electricity demand for cooling and degrading the performance of generation, transmission, and distribution infrastructure. Elevated ambient temperatures increase conductor resistance and cause thermal sag, forcing derating of line ampacity. These coupled stresses tighten reserve margins, heighten congestion, and elevate the risk of outages, with empirical data showing increases in outage frequency by approximately 4% and duration by 8% during heatwave periods. Data-driven optimization offers a powerful pathway for mitigation by leveraging high-resolution weather forecasts, historical outage records, SCADA/PMU telemetry, and climate projections to inform adaptive decision-making in unit commitment, economic dispatch, and transmission operations. This research paper provides a comprehensive assessment of the impact of heatwaves on power system reliability and explores mitigation strategies through data-driven optimization frameworks, including stochastic programming, distributionally robust optimization (DRO), and hybrid machine learning-enhanced models. It formulates temperature-dependent mixed-integer linear programs (MILP) that incorporate dynamic line ratings (DLR), temperature-adjusted loads and generation limits, and equity-aware objectives. Case studies from major heat events demonstrate that data-driven approaches can reduce expected unserved energy by 30–60%, lower operational costs, unlock additional transmission capacity via predictive DLR, and improve equity outcomes by protecting vulnerable communities. As heatwaves become more frequent and severe, integrating data-driven optimization into operational and planning practices is essential for enhancing grid resilience, maintaining reliability, and ensuring equitable energy access under increasing thermal stress.