@article{Rong2025, 
author = {Xuanman Rong and Yuxiong Huang and Gengfeng Li and Zhaohong Bie},
title = {Generative AI-augmented Monte Carlo Simulation for Resilience Assessment of Typhoon-affected Power Distribution Systems},
year = {2025},
journal = {CSEE Journal of Power and Energy Systems},
volume = {11},
number = {6},
pages = {2659-2672},
keywords = {Active distribution system, extreme event, generative adversarial networks, machine learning, Monte Carlo simulation, resilience assessment, typhoon},
url = {https://www.sciopen.com/article/10.17775/CSEEJPES.2025.06160},
doi = {10.17775/CSEEJPES.2025.06160},
abstract = {The increasingly frequent extreme events pose a serious threat to the safe operation of distribution systems. A detailed description of the impact of extreme events on the supply-network-demand status of distribution networks is crucial for scientific disaster prevention of power systems. To this end, this paper proposes a method for generating disaster scenarios by combining conditional generative adversarial networks (CGAN) with Monte Carlo simulation (MCS) and further proposes a quantitative assessment method for the resilience of distribution systems under typhoon conditions. First, an event-triggered resilience assessment framework combining Monte Carlo simulation and generative AI-driven techniques is proposed. Next, a simulation-based disaster scenario-generation algorithm that considers spatiotemporal correlated supply-demand uncertainty and typhoon-affected component vulnerability is developed. Then, a series of event-affected resilience indices are defined, and the impact of a typhoon on distribution network performance is calculated by simulating during-event disaster scenarios and performing post-event restoration strategies. Finally, extensive numerical results validate the effectiveness of our proposed method.}
}