A Resilience Assessment Framework for Distribution Systems Under Typhoon Disasters

Yuantao Wang, Tianen Huang, Xiang Li, Jian Tang, Zhenjie Wu, Yajun Mo, Lin Xue, Yixi Zhou, Tao Niu, Sicong SunIEEE Access 2021 · DOI

Introduction

As climate change intensifies the frequency and severity of extreme weather events, power distribution grids face unprecedented threats from typhoons. Traditional grid reliability metrics often rely on the "N-1" criterion—the system's ability to withstand the loss of a single component. However, typhoons present a different challenge: widespread, simultaneous failures across a geographic area. Current planning tools often struggle to account for the inherent uncertainty of a typhoon’s path and the spatiotemporal nature of the disaster—how the storm’s impact evolves as it moves across the grid infrastructure.

The Proposed Framework

This research introduces a comprehensive framework designed to quantify grid resilience by modeling the interaction between probabilistic storm behavior and grid vulnerability. The framework is structured into three primary modules that move from environmental modeling to operational impact assessment.

Overview of the proposed resilience assessment framework, detailing the progression from typhoon modeling to grid vulnerability and final resilience assessment.

1. Probabilistic Typhoon Modeling

Rather than analyzing a single historical storm, the framework models four core typhoon parameters: landing location, central pressure difference, movement speed, and direction. By applying log-normal and paranormal probability distributions to these variables, the researchers generate 81 distinct, realistic typhoon scenarios. This probabilistic approach ensures that grid planners can prepare for a wide range of potential impacts rather than just the "worst-case" historical event.

2. Spatiotemporal Vulnerability Modeling

The framework establishes failure rate models for both poles and lines. Poles are modeled using a segmental function where failure occurs once wind speeds exceed a specific threshold. Lines follow an exponential failure curve based on both wind speed and the duration of exposure.

Spatiotemporal interaction between a moving typhoon wind field and distribution grid infrastructure.

As shown in the figure above, the impact is dynamic. A line's vulnerability changes as the typhoon's center ($O$) and its maximum wind field radius ($r_{max}$) move along a track. This spatiotemporal modeling captures the reality that a line safe at the start of a storm may reach peak risk hours later as the storm progresses.

3. Islanding and Optimization

When lines fail, the grid often "splinters" into isolated sections. The framework utilizes a Breadth-First Search (BFS) algorithm on the grid’s incidence matrix to identify these isolated "islands" in real-time. For each identified microgrid, a linear optimization model (implemented via YALMIP and CPLEX) is used to minimize load shedding. The final output is a Resilience Index, calculated as the product of a typhoon scenario's probability and the resulting optimal load shedding.

Case Study and Results

The framework was validated using the IEEE 33-bus test system mapped onto the geography of Zhejiang Province, China. Key findings include:

  • Impact of Distributed Energy Resources (DERs): The study demonstrates that DERs (solar, wind, and batteries) are critical for resilience. In one test zone, the Resilience Index (expected load shedding) dropped from $1.26 \times 10^{-7}$ MW to $8.66 \times 10^{-8}$ MW when DERs were present.
  • The Role of Storage: Energy Storage Devices (ESDs) act as a vital buffer, providing the necessary time for the system to stabilize during the most volatile periods of the storm when generation and load are highly mismatched.
  • Targeted Hardening: By identifying "weak lines" that consistently failed across multiple probabilistic scenarios, the framework provides a data-driven roadmap for physical upgrades, such as undergrounding lines or reinforcing specific poles.

Conclusion and Future Work

This framework moves grid assessment from qualitative "gut feelings" to a quantifiable Resilience Index. While robust, the authors note that future iterations will aim to include cascading failure models—where the electrical stress from one failed line triggers others—and restoration modeling to account for the speed of repair crews in post-storm conditions.

Summary updated

BibTeX
@article{wang2021resilience,
  author = {Yuantao Wang and Tianen Huang and Xiang Li and Jian Tang and Zhenjie Wu and Yajun Mo and Lin Xue and Yixi Zhou and Tao Niu and Sicong Sun},
  title = {{A Resilience Assessment Framework for Distribution Systems Under Typhoon Disasters}},
  journal = {IEEE Access},
  year = {2021},
  doi = {10.1109/access.2021.3128967},
}
Figure