Why wind forecasts might be the fix for grid congestion
A conductor's heat limit, not a lack of wires, is often behind grid congestion. Meet the company forecasting its way around it.
Published on September 16, 2026
.png&w=2048&q=75)
© Gridraven
Mauro swapped Sardinia for Eindhoven and has been an IO+ editor for 3 years. As a GREEN+ expert, he covers the energy transition with data-driven stories.
Grid congestion sounds like an infrastructure problem — too few wires, not enough capacity. But according to Georg Rute, CEO of Estonian grid intelligence company Gridraven, the real bottleneck is often narrower. “The underlying problem is a lack of transmission capacity, but physically, the real limit is conductor overheating,” he says.
Power lines can safely reach 75 to 90°C — a ceiling that's easy to hit on a hot, still day, and barely in play on a windy one. Predicting which scenario a line will face tomorrow, or next week, is where Gridraven has built its business.
The company focuses on accurately forecasting wind around power lines so grid operators can trust their grid capacity forecasts days in advance. Thermal limits account for, by Rute’s estimate, roughly half to two-thirds of grid congestion.
Beyond Borders
Innovation doesn’t stop at the border! In Beyond Borders, we spotlight disruptive innovations worldwide that can help tackle societal challenges in the Netherlands. How are they making an impact? How can they help?
Weather’s impact on power lines
Renewable energy is often generated far from where it is consumed, and in peak production times, lines might not have enough capacity to carry power from an offshore wind farm to a city center hundreds of kilometers away. What happens is that windmills or solar panels are then turned off – curtailment in energy jargon – meaning clean electricity goes to waste.
“If it's cold, cloudy, or especially if it's windy, the line can handle more. Day versus night can be worth about 10% more capacity. Winter versus summer — say 0°C versus 20–30°C — might give you another 15–20% from better cooling. But a light wind can roughly double the line's capacity. It works the same way a breeze cools you down — it's just very effective cooling,” explains Rute.
Historically, when building a new line, operators designed for the worst-case scenario: a hot summer day with no wind. They made sure the line could carry the required power under that condition. Building variable generators in remote locations puts additional pressure on the grid: we suddenly need to use that extra capacity that has always existed, but to do so safely, we need an accurate forecast.
An algorithm to forecast weather conditions
What Gridraven has built is a sensorless version of a technology called dynamic line rating (DLR). Essentially, DLR means using actual or forecast weather conditions to determine how much power a line can safely transport. Many other companies do something similar by installing sensors on conductors. But sensors measure, not predict, and, according to Gridraven’s CEO, what makes a difference in DLR is weather-forecasting capability.
For this reason, the Estonian company skipped the hardware and focused solely on forecasting. The company layers its own machine learning algorithms on top of standard weather forecasting — Germany’s ICON-EU model, for example — using it to downscale regional forecasts to a single point on a power line.
It does this for every "span," the stretch between two towers, focusing on the midpoint, which is typically the lowest and most wind-sheltered part of the line. A typical line might be broken into around 300 forecast points.
Gridraven engineers trained the model using two main data streams: LiDAR elevation data and historical weather data from thousands of stations worldwide. In particular, LiDAR maps, widely available and cheaper than satellite imagery, allow mapping at resolutions up to 1 meter. From these two sources, the model learns how terrain bends and slows the wind. The output is a single number, which goes straight into a grid control room and tells how much load a line can safely carry tomorrow and up to ten days out.
Gridraven's CEO Georg Rute - © Gridraven
How Gridraven got here
Rute isn't an outsider to this world. He worked for years at an Estonian national grid operator, where the company tried installing weather stations on towers to see if forecasts could be used to set more accurate line ratings. It didn't work. He then moved to a company working on electrolyzers, where solar and wind developers he spoke with worldwide voiced a recurring complaint: there are plenty of sites to build, but no way to connect to the grid.
These two problems — unreliable forecasts and a grid too conservative to use its own spare capacity — became the basis for Gridraven. Three years ago, he reached out to his former colleague Henri Manninen, a power-systems PhD, and AI specialist Markus Lippus joined soon after. The company now employs 20 people.
Gridraven’s impact
Gridraven is live across Finland's full grid and part of Estonia's. Rute pushes back on the industry's reputation for being slow to adopt new technology. "I don't think that's true," he says. "I think it's a bit of a lazy excuse that solutions providers make when their solution doesn't actually work."
In his experience, it's the opposite: "As soon as we're able to show that it really works, there's a lot of interest happening in parallel." Operators, he says, simply can't afford technical risk on anything unproven.
Rute says Gridraven's forecasts unlock roughly 20–40% more theoretical line capacity. These translate into about 10% more usable capacity in energy markets once other constraints are factored in. The figure was validated in a study conducted in Finland, while a separate study in Estonia found a 22% increase.
Because wind and solar generation often coincide, the extra capacity disproportionately benefits renewables. Rute also points to a knock-on effect on prices: much of the price gap between neighboring countries comes down to insufficient transmission between them, and more capacity can mean, for example, Germany relying less on coal and importing more nuclear power from France. Gas plants — which typically cover supply shortfalls — stand to lose the most from that shift.
Time to scale
For now, Gridraven considers its core technology mature and is focused on incremental refinements — adding cloudiness forecasts, for instance, which can be worth another 5% of capacity on an overcast day — while it shifts attention to scaling up. "I don't think transmission utilities are slow," Rute says. "I think everybody in transmission has realized that they need to act very fast." The next test for the company is whether the rest of the industry agrees fast enough to keep up.