Risk Assessment of Transmission Lines Against Grid-Ignited Wildfires

Wildfires ignited by the power lines have become increasingly common over the past decade. Enhancing the operational and financial resilience of power grids against wildfires involves a multifaceted approach. Key proactive measures include meticulous vegetation management, strategic grid hardening such as infrastructure undergrounding, preemptive de-energization, and disaster risk financing, among others. Each measure should be tailored to prioritize efforts in mitigating the consequences of wildfires. This paper proposes a transmission line risk assessment method for grid-ignited wildfires, identifying the transmission lines that could potentially lead to damage to the natural and built environment and to other transmission lines if igniting a wildfire. Grid, meteorological, and topological datasets are combined to enable a comprehensive analysis. Numerical analysis on the standard IEEE 30-bus system demonstrates the effectiveness of the proposed method.

💡 Research Summary

The paper addresses the growing problem of wildfires ignited by electric transmission lines and proposes a comprehensive risk‑assessment framework that integrates landscape, meteorological, and grid topology data. Traditional wildfire‑risk tools (e.g., FHZS, USFS Wildfire Hazard Potential, WiNGS) focus on climate, vegetation, and topography but do not explicitly consider the electrical infrastructure as a fire ignition source. To fill this gap, the authors develop a four‑step methodology.

First, they collect eight landscape variables (elevation, slope, aspect, fuel type, tree‑cover fraction, tree height, canopy‑bottom height, canopy density) together with weather parameters (wind direction, wind speed, temperature, humidity) and a detailed transmission‑network model that provides latitude/longitude for every line segment. For each transmission line, a set of ignition points is generated (three per line in the case study).

Second, the ignition points and the assembled data are fed into the FARSITE fire‑behavior simulator, a deterministic model that predicts fire spread based on the supplied terrain, fuel, and weather inputs. The output is a spatial fire‑perimeter for each ignition scenario.

Third, the burned area is intersected with the landscape map to compute total acreage, and with the grid topology to identify which lines are intersected by the fire. Two cost components are then calculated: (a) environmental loss = burned acres × average cost per acre (CBE) and (b) line‑reconstruction loss = total length of damaged lines × average reconstruction cost per mile (CBL). The sum yields a total financial loss L j for line j.

Fourth, a normalized risk metric M j = L j / L max is defined, where L max is the worst‑case loss among all lines. M j ranges from 0 (no risk) to 1 (maximum risk) and provides a quantitative ranking of transmission lines for prioritizing mitigation actions such as vegetation management, undergrounding, or pre‑emptive power shutoffs (PSPS).

The methodology is demonstrated on the IEEE‑30‑bus test system, which contains 41 branches (7 transformers, 34 lines) representing a network over the western edge of Yosemite National Park and Stanislaus National Forest in California. Weather data for all four 2022 seasons are used, and three ignition points per line generate 408 simulation scenarios.

Results show strong seasonal variation. Line 6 has the highest average burned area (≈4 236 acres per year), making it the most environmentally hazardous line, especially in summer. Line 3, although not the longest, causes the greatest grid damage—an average of 259.9 miles of line must be rebuilt—because it is long and densely surrounded by other lines, leading to cascading failures. Winter risk concentrates on lines 35‑38, reflecting the influence of lower humidity and different wind patterns.

The proposed risk metric successfully isolates the lines that pose the greatest combined environmental and grid risk, offering utilities a data‑driven tool for allocating limited resources. The authors acknowledge limitations: FARSITE’s deterministic nature may under‑represent uncertainty in wind shifts and fuel moisture; ignition point selection is heuristic rather than based on actual fault data; and cost parameters (CBE, CBL) can vary regionally and temporally.

Future work is suggested to incorporate probabilistic fire‑spread models, real‑time weather feeds, fault‑based ignition modeling, and multi‑objective optimization that includes societal and health impacts alongside financial losses. Overall, the paper contributes a novel, integrative approach to quantifying transmission‑line‑initiated wildfire risk and demonstrates its practical applicability on a standard test system.