Compound Climate Risk-Aware Electricity Dispatch: Trajectory-Based Modeling for Power System Resilience

Abstract

In recent decades, emerging compound hazards in a changing climate have increasingly threatened power system resilience and posed new challenges to electricity dispatch. Conventional Value-at-Risk (VaR)-based risk-aware dispatch provides a tractable means of managing stochastic net demand deviations caused by distribution feeder outages. However, when applied to temporally compounding events such as tropical cyclone (TC)--blackout--heatwave, conventional net demand risk formulations may assign similar risk levels to a brief outage and an outage with the same instantaneous net demand impact that persists into a subsequent hazard, despite their different requirements for sustained operational flexibility. Moreover, directly incorporating compound risk into electricity dispatch models faces a structural mismatch: the risk is defined by the joint distribution of feeder restoration and subsequent hazards over exponentially many joint states, whereas dispatch models require risk formulations that are tractable in the scheduling decisions. To address these issues, this paper proposes a compound climate risk-aware dispatch framework with a trajectory-based compound risk model, which defines the loss on operating deviations determined by entire feeder connection trajectories, parameterized by uncertain restoration times and coupled with subsequent hazards. Analytical moment-generating functions and equivalent exponential-cone reformulations provide a compact representation of the resulting risk constraints, preserving the relevant dependencies without explicit joint-state enumeration. By leveraging a physics-based spatiotemporal TC--blackout--heatwave hazard model, case studies demonstrate the effectiveness of the proposed method with real-world time-series power outage data from Puerto Rico, including data collected during Hurricane Fiona in 2022.