Storm-proofing 1% of power lines protects entire grid from blackouts

Storm-proofing as little as 1 per cent of the power lines in an electricity grid could slash the chance of hurricane-induced blackouts by between fivefold and 20-fold, a simulation suggests. The demonstration, which took place in a simulated version of the Texas electricity grid, could help boost the resilience of power transmission systems worldwide.

“The importance of various lines to the overall system only becomes apparent if we study the partially destroyed states of the grid that occur as the storm progresses,” says Frank Hellmann at the Potsdam Institute for Climate Impact Research in Germany.

To identify those critical power transmission lines most in need of protection, Hellmann and his colleagues examined how the grid responds to widespread damage over time. They focused on large “failure cascades” that occur after the initial storm damage: as power plants and transmission lines shut off to protect themselves from additional damage, they cause secondary power outages that can broaden the hurricane’s impact.

The researchers simulated both wind-related storm damage – such as gusts damaging towers or taking down tree branches that fall onto transmission lines – and the resulting cascade of power outages that occurred in the Texas power grid during seven historical hurricanes between 2003 and 2020.

Instead of trying to predict individual power line failures that can occur from a fallen tree or a lightning strike, the researchers assigned each line a probability of failure based on local wind speeds during each storm event. Their model consistently identified the same 20 critical lines where initial storm damage could trigger a cascade of secondary line failures – even when they reran the simulation with random variations in each line’s probability of failure.

The experiment relied on a synthetic network model of the Texas grid previously developed by a Texas A&M University team. It represents the grid’s overall behaviour without being an exact replica of the actual physical grid. “None of the transmission lines in that grid are real lines,” says Adam Birchfield at Texas A&M University. “So to find out whether these results are valid for the real Texas grid, at a minimum the study would need to be run on a model of the real Texas grid.”

Although independent researchers typically lack access to such models for security reasons, the power grid operators themselves could run this simulation on their own detailed grid models. Once they figure out which specific lines are weak points, they can weatherproof those crucial components of the grid.

Beyond Texas, such simulations could also model the grids of other locations that experience similar storm events. That “may offer opportunities to verify the model and results”, says Chuanyi Ji at the Georgia Institute of Technology in Atlanta, who was not involved in the study.

The model of wind-related damage has its limits, acknowledges Hellmann. It does not account for additional possible damage from flooding, or for how power grid operators can take precautionary measures to prevent power outages.

Still, the study’s main takeaway is reinforced by having used a “wide variety of scenarios” to check the power failure probabilities in a realistic grid model, says Birchfield. “I do think that hardening transmission corridors is an important component of increasing electric grid resilience,” he says. “And the paper demonstrates that a strategic choosing of transmission lines to harden is important to having the biggest impact on resilience.”