Using Utility Vegetation Management Software for Wildfire Mitigation

We’ve talked about GridOS® Visual Intelligence on this blog for its value in utility vegetation management (UVM), and more recently as a tool for assessing storm damages.

Today I wanted to talk about another Visual Intelligence use case: wildfire mitigation.

Wildfires have always posed a risk to electric grids, although in recent years their frequency and severity has increased. This existential threat is forcing utilities to mitigate wildfire risk with multiple approaches. In reviewing various utilities’ Wildfire Mitigation Plans (WMPs) , four common themes are evident:

Traditionally, news footage has shown wildfires strictly in very rural areas. In recent years, however, wildfires have been observed spreading into more urban, heavily populated areas as well such as Maui or Los Angeles.

The Los Angeles wildfires were especially disturbing, given that one of the most affected areas was a densely packed metropolitan area. Office buildings, apartments, front lawns, and shopping centers went up in flames – as did all the grid infrastructure serving them. These were not the rural wildfires we were used to seeing – these were complex infernos spreading into urban areas. For utilities, this evolution of the "wildfire" represents a collapse of the traditional wildland-urban interface. The challenge is no longer just protecting transmission lines in the forest; it now includes maintaining the structural integrity and fire-readiness of the distribution network.Anecdotally, wildfire risk was a major topic of conversation at GE Vernova’s Orchestrate 2025. Many utility professionals spoke of planned investments in solutions that identify wildfire-prone areas and prevent ignitions. Even utilities operating in states not widely known for wildfires expressed heightened anxieties about a wildfire impacting their operations and causing a major power grid disruption.

Visual Intelligence is an invaluable solution for wildfire mitigation. And here’s how.

As explained in past blogs, Visual Intelligence is perhaps best-known as a proven UVM solution, increasing the efficiency, effectiveness, and precision of vegetation management activities. Visual Intelligence shows utilities and vegetation service providers the exact locations and conditions of encroaching hazard trees and the fall-in potential of vegetation that could cause lengthy power grid disruptions.

Utilities already using Visual Intelligence to detect encroachment can also identify wildfire risks without doing anything differently. The compiled vegetation data can be overlayed with risk models, fuel levels, weather maps, and additional wildfire mitigation specifications. The findings and outputs on areas prone to wildfire risk are revealed with a simple, easy-to-use color-coding scheme for end users.

For example, look at the below Visual Intelligence screenshot, showing a length of rural power lines using LiDAR data sources:

With the help of this visual, trimming crews can instantly see where trees are encroaching on the power lines, or where vegetation has growth within the required space around the poles (marked by the white circles). A simple white box marks the work zone for the trimmers’ reference upon arrival at the site. Intervention will be faster, easier, and more accurate, thus both preventing wildfire ignitions and protecting the assets from any spreading wildfires.

Or consider this satellite image, which was fed into Visual Intelligence to assess the health of the pictured trees:

Visual Intelligence’s built-in analytics engine can assess vegetation height and tree health to determine whether any trees are in danger of falling and igniting or fueling a blaze. In the image, the color coding of each individual tree indicates their heath and risk of becoming kindling in a future wildfire, with red trees being the highest risk. The application also calculates and models vegetation quantity to determine exactly how much a crew should trim and clear to best protect assets. That is evident in the next image, depicting another Visual Intelligence screenshot of vegetation growing around and beneath a set of transmission towers.

There are several areas of interest inthis image. At bottom right, the foliage of a tall tree is colored blue, indicating protruding branches that are getting dangerously close to the power lines and the towers supporting them. Meanwhile, at center left, more vegetation is colored blue – but this vegetation is clearly not as tall as the problematic tree at right. This coloring indicates overgrown bushes and shrubbery that could fuel a passing wildfire. At center right, the yellow circle shows vegetation encroaching the utility’s required defensible space around pole. Ensuring this zone is clear of vegetation adds further reassurance of protection to the transmission structure from a passing wildfire.

Though many wildfires are ignited by natural phenomena like lightning, or human-caused issues like poorly extinguished campfires, there are factors utilities can control to minimize the likelihood of a wildfire ignition.

That’s where Visual Intelligence comes in again. Visual Intelligence is also used to identify equipment defects and other physical damages to grid assets that could potentially ignite a wildfire.

For example, frayed wires can release sparks if the conditions are just right. Visual Intelligence can identify fraying, damaged wires and alert the appropriate maintenance personnel.

Or as another example, think about lattice transmission towers. They are typically made of galvanized steel – “galvanized” meaning that the steel is coated with zinc to protect it from rust and corrosion. But galvanization is not permanent. The combination of wind, direct sunlight, and precipitation can wear off the zinc layer over time. Even the smallest scratch can expose raw steel to the elements, resulting in corrosion and rust. Without intervention, the corrosion can eventually compromise structural integrity and failure of electrical components. And it goes without saying that a transmission tower collapsing, a broken crossarm, or even just a broken bolt caused by corrosion can be the root cause of a wildfire ignition.

This level of risk exposure is why some of GE Vernova's biggest customers are using Visual Intelligence to monitor their transmission towers and distribution poles for signs of corrosion, cracks, broken insulators, and any other physical damages. And with its streamlined workflow and automation capabilities, Visual Intelligence can identify and eliminate wildfire risk factors before they cause a spark. The application makes an immediate impact as utilities push toward ever-greater resilience and risk reduction.

Now that we’ve covered Visual Intelligence’s workflows for identifying wildfire risk, the next question is how to use the solution’s capabilities to unlock predictive risk mitigation – in other words, the ability to proactively identify and address wildfire risk factors before they can damage assets.

Moving from reactive to proactive workflows is easy and seamless with Visual Intelligence.

Phase 1: Wide-Area Resilience Scanning

The process begins by eliminating information gaps across the entire service territory, including inaccessible rural spans. Rather than relying on subjective ground patrols, Visual Intelligence establishes an objective baseline using LiDAR, satellite, and ground-based imagery to identify macro-level anomalies and high-consequence "red zones" where environmental conditions, asset density, and asset health converge to create maximum risk.

Phase 2: Deterministic Anomaly Detection & The Visual Health Index (VHI)

Visual Intelligence processes multi-modal imagery to move beyond binary "Pass/Fail" inspections. The system generates a visual health index for every asset, mathematically proving asset health based on its recognition of specific physical indicators of risk, including cracked insulators, corroded cross-arms, and beyond. This empowers asset managers to execute precise, data-guided “keep/replace" decisions about their assets. It also helps them prioritize or defer unnecessary OpEx spend on healthy assets.

It all adds up to a streamlined process for proactively mitigating the risk of wildfires across the grid, thanks to the visual insights and predictive workflows of Visual Intelligence.

For more information on GridOS Visual Intelligence and its many use cases, check out our solution paper on the subject, “Orchestrate the Grid with Visual Precision: GridOS Visual Intelligence.”