Gridraven’s journey began when we noticed energy providers worldwide struggling to secure grid connections. It takes 10 years on average to build a new high-voltage line — time that we simply can’t afford, as it slows down economic growth. Our solution increases grid capacity by 30% on average and requires zero hardware.
The existing grid isn’t fully utilized because power lines, capable of carrying up to four times more power under cold or windy conditions, are operated conservatively due to a lack of precise weather data. Our technology changes that. Integrating weather data will boost grid capacity by a third annually, accelerating affordable energy by a decade. The key is to make this additional capacity available for day-ahead energy markets so that the lowest-cost power plants can be dispatched.
The lack of affordable energy is the biggest obstacle to economic growth.
THE REASON FROM VISION TO REALITY
Using machine learning to add more capacity to energy markets.
CONTENTS
SHAPING THE FUTURE OF ENERGY
The importance of collaboration and innovation in reshaping the global energy landscape.
7 10 16 26 8 14 20 28
THE FOUNDERS
The driving force behind Gridraven: Meet the founders.
THE IMPACT OF DLR
The grid needs an upgrade due to the growing demand for power.
PILOT: ELERING
We provide accurate and reliable forecasts even in challenging conditions where traditional methods fall short.
PREDICTING CALIFORNIA’S WINDS
Our machine learning model improves wind speed forecast accuracy in the Golden State by 50% compared to existing largescale numerical weather predictions.
PROBLEM + SOLUTION
What are the problems in the energy sector, and how can they be solved.
The reason we need to talk about power grids
At Gridraven, our mission is to unlock the untapped potential of power grids. The lack of affordable energy is the biggest obstacle to economic growth, and Gridraven is on its way to solving this. By using cutting-edge machine learning, our system optimizes the grid’s capacity based on actual weather conditions with meter-scale accuracy. Power lines can carry much more energy than they’re currently allowed to—if it’s cooler or windier, they can handle more load.
Competing solutions rely on hardware that is good for real-time measurements, but cannot be used for forecasting. Our technology unlocks more capacity for day-ahead energy markets, helping to lower energy costs and accelerate growth.
Our team combines decades of experience in power systems, machine learning, and energy solutions. We’re here to make grids more efficient, reliable, and ready for the future. We understand the challenges firsthand and are ready to provide the push needed to make progress easier and faster.
Georg Rute CEO
the world with abundant energy
Powering
Sensorless Dynamic Line Ratings
The grid needs an upgrade as they are under increasing stress due to the growing demand for power around the world.
Software only
Gridraven’s software-only solution determines the detailed conditions across all power lines, enabling operators to increase capacity confidently.
+30% more capacity
DLR technology boosts power transmission capacity by 30% without new infrastructure.
Forecast 48 hours ahead
Energy markets and dispatch decisions are made hours and days in advance. Forecasting line ratings helps power plants produce more power.
Hyper-local weather
Gridraven applies machine learning and AI to make extremely detailed weather models based on which grid capacity is adjusted dynamically.
Lower energy prices
More cheap energy Ready now
Problem
to build a new power line. This is slowing down the growth of industry.
1500 GW of projects that are waiting for a grid connection
Our solution from the existing grid. Now. +30% more
Removing congestion
Load growth and the increase in solar and wind are driving congestion in grids globally. Additional capacity helps remove congestion. Serving load growth
More energy is necessary to support economic growth.
From Vision to Reality:
How Gridraven is Rethinking Power Grid
Potential with AI
As the demand for energy grows, so does the need for a power grid capable of adapting to unpredictable weather and increased load without extensive new infrastructure. Markus Lippus and the team at Gridraven are rethinking grid potential through advanced AI solutions, using machine learning to unlock real-time flexibility in power transmission. By integrating environmental data with predictive models, they’re enabling the grid to safely handle more power when conditions allow—helping accelerate the shift toward sustainable energy while reducing costs and congestion.
Building Practical AI Solutions
As a Data Scientist, AI consultant, and Machine Learning Solutions Architect, I’ve built AI systems that help people collect and process information, directly speak to them or quietly work in the background, reaming tons of data to categorize it, detect faults on production lines or flag potential taxevaders. There are as many use cases for AI as there are tedious tasks people have to do and then some, and while the apparent ecological impact of the current hype around LLM-s troubles me, I see the good it can do along with the dangers of misuse.
AI at Gridraven: Optimizing for the Environment
Now, at Gridraven, I’m far from the directly visible part of AI and, maybe ironically, trying to make it work to benefit the environment. Specifically, we’re trying to address the everincreasing need for electricity and the bottlenecks in transmitting it without
building new infrastructure and allowing more renewables on the grid. The way to do it is by reducing inefficiencies in the existing grid by enabling it to adapt to environmental conditions. One of the things limiting power transmission is that the wires used for it get hot and break down - more power gets you more heat, so strict limits are in place. However, weather can have quite a significant effect on this, both by cooling and heating the wires. Theoretically, it is substantial enough to increase the transmission limits by up to 3x in some cases. For a number of good reasons and a few bad ones, this and the changing nature of weather is largely ignored.
Adapting Existing Research for Gridraven’s Unique Needs
As there’s little to grab and use off the shelf, our work is about finding the bits and pieces from related research that show potential in related uses and adapting these to our rather specific
case. Luckily, what we don’t lack is data - there are decades of remote sensing data from satellites packed with sensors, weather simulations, and historical forecasts, observation data from the polar bears to penguins, etc. There’s also more and more highly detailed data, mapping the earth at <1m resolution. The algorithms used for modeling the climate and weather can’t even use this level of detail not only because they’re not designed to take advantage of the information, but the scale of computation itself would be prohibitive.
Markus Lippus Chief Data Scientist
Teaching AI to Understand Environmental Interactions
Our algorithms tackle the challenge of understanding how every tree, building, and terrain feature influences weather just above the ground. Unlike humans, AI lacks “common sense” and starts without a model of the world, so we teach it how vegetation, terrain,
and atmosphere interact. Using selfsupervised learning, the AI extracts patterns, learning how forests block wind or how vegetation changes wind direction. This process builds a form of environmental “intuition.”
However, real-world data is messy— forests get cut, buildings emerge, and weather stations provide inconsistent measurements. Our models adapt to these inaccuracies, flagging when predictions may be less reliable and highlighting areas needing better data. We combine methods from climate modeling, deforestation monitoring, and remote sensing into a probabilistic neural network. Instead of rigid predictions, it offers ranges like “almost certainly at least 3m/s wind,” with confidence intervals that help make informed, risk-aware decisions.
Building Smarter, More Adaptable Grids
GridRaven’s mission is to transform the way we use the power grid, making it smarter and more adaptable. By applying AI to dynamically optimize transmission capacity, we’re addressing current energy needs while laying the groundwork for a resilient, sustainable future. Our technology empowers grids to work harder, safer, and cleaner— unlocking their full potential.
Advancing Grid Capacity and Forecasting with Elering:
Pilot results
We signed a pilot project agreement with Elering to implement our innovative softwarebased solution, covering 5,000 km of Estonia’s electricity grid and encompassing a total of 15,000 spans. The aim of the pilot collaboration is to increase the capacity of existing lines and improve forecasting accuracy under complex conditions.
Our solution demonstrated significant success:
Average Forecast Accuracy
Reliability
Capacity Increase
60% better than existing forecasts.
The 95% confidence interval was maintained 95.1% of the time.
During the period of September–November, Gridraven’s forecasted line capacity was a third higher than the capacity achieved with the previously used approach, while still ensuring that confidence intervals for constantly changing variables were met.
The initial results of the pilot project are in and are based on tests conducted on a 110 kV line, where forecasts were validated against eight measurement points up to 48 hours ahead. The trials took place in a forested area where current wind forecasts had previously failed to provide reliable results.
Innovation in Forested and Complex Terrain
The pilot proved that Gridraven’s solution can provide accurate and reliable forecasts even in challenging conditions where traditional methods fall short. This opens up the possibility of increasing the utilization of the existing electricity grid without additional infrastructure investments, enabling the integration of more renewable energy sources into the grid.
Elering
Elering is Estonia’s national transmission system operator (TSO) responsible for managing and operating the country’s electricity and gas transmission networks. It plays a critical role in ensuring a secure and reliable energy supply, balancing electricity demand and production, and maintaining the infrastructure required for energy transmission.
Shaping
the Future of Energy:
Henri Manninen’s Journey Through Innovation and Expertise
We sat down with Gridraven’s Co-Founder and CTO Henri Manninen, PhD, whose story underscores the importance of interdisciplinary collaboration and innovation in reshaping the global energy landscape.
What inspired you to co-found Gridraven?
While working at Elering and conducting research during my PhD, I saw how conservative and inefficient the energy sector can be. There’s a saying among engineers: “The physics of electricity hasn’t changed in 60 years.” That’s true, but advancements in technology, such as machine learning, now allow us to implement methodologies that were once impossible or economically unfeasible.
Dynamic Line Rating (DLR) is a great example. While the concept has existed since 1958, technology has only recently advanced enough to make it practical and impactful.
How has your expertise contributed to your work at Gridraven?
My seven years in the Estonian transmission system gave me a deep understanding of how power systems, particularly transmission lines,
work in practice. My PhD research, which combined theoretical and interdisciplinary approaches, taught me the value of integrating diverse fields to solve complex problems. This unique blend of practical experience and academic research has laid the foundation for developing innovative solutions at Gridraven.
What excites you most about working at the intersection of technology, machine learning, and energy systems?
No one is an expert in everything, and solving interdisciplinary problems requires collaboration with specialists in diverse fields. Working with brilliant people to tackle complex challenges makes the journey incredibly exciting and educational.
What specific problems are you solving at Gridraven, and why are they important for the energy industry?
We’re addressing the limited capacity of existing transmission lines.
Transmission lines are the backbone of energy systems, and upgrading them to accommodate new generation or consumption units is expensive and time-consuming.
Our solution increases the capacity of overhead lines by up to 30% annually— sometimes as much as 50% during windy periods—without requiring new infrastructure. This can significantly alleviate bottlenecks, reduce costs, and expedite grid expansion efforts.
Henri Manninen PhD, CTO
Can you walk us through a recent project you’ve worked on? What were the main challenges and outcomes?
We’re currently piloting our solution with Elering, covering 5,500 km of transmission lines. This project involves validating predictions of line ratings based on weather and temperature data.
Our initial results are promising, showing we can predict line ratings up to 48 hours in advance with predetermined confidence intervals. Over a two-month period, we achieved an average capacity increase of nearly 40% compared to traditional approaches, using a 95% confidence level.
The biggest challenge has been managing the sheer volume of data— around 15,000 line rating predictions per hour across the Estonian grid. However, seeing the results align with measurements and exceed expectations has made the effort worthwhile.
Can you share an example of a technical breakthrough or innovative approach you’ve introduced at Gridraven?
Our biggest breakthrough is determining dynamic line ratings without relying on hardware. Most competitors use sensors on highvoltage lines, but our approach relies purely on data. By accurately predicting cooling and heating inputs, we can estimate line ratings effectively, even accounting for random variables like weather conditions.
What’s one surprising insight or result you’ve uncovered in your work so far?
We’ve demonstrated the ability to predict safe, operable line ratings several days in advance with confidence intervals that hold true. This opens up significant opportunities for the dayahead electricity market, where our predictions could reduce prices and improve grid efficiency.
How do you see the energy sector evolving with the integration of machine learning and advanced technology?
The energy sector will always be conservative due to its critical role in society. However, utilities are increasingly open to adopting AI and ML as they recognize the potential benefits. The key is ensuring these technologies are rigorously tested and proven before implementation.
What role does Gridraven play in this transformation, and what makes its solutions unique?
Our solution leverages advancements in technology and machine learning to unlock the full potential of existing infrastructure. Unlike traditional methods that require costly hardware or supercomputers, we deliver results efficiently and reliably using state-ofthe-art algorithms.
What potential long-term impact do you hope your work will have on energy systems and sustainability?
Our goal is to maximize the capacity and safety of overhead lines, improving the efficiency of existing infrastructure and reducing bottlenecks in electricity markets. This can lower energy costs and support the transition to more sustainable energy systems.
What’s one thing about your work at Gridraven that makes you proud?
Building an incredible team and turning an ambitious idea into something meaningful and impactful.
What message would you like to share about the importance of innovation in the energy sector?
Power systems are largely built on decades-old principles. With digitalization and advancements in machine learning, we now have the tools to implement solutions that were once purely theoretical. While we shouldn’t rush to apply AI everywhere, it’s essential to carefully explore its capabilities and potential.
Predicting California’s Winds:
Hyperlocal Machine Learning Takes Forecasting to New Heights
Gridraven’s machine learning model improves wind speed forecast accuracy by 50% in the Golden State compared to the existing large scale numerical weather predictions. These improvements are fully leveraged in our software-based Dynamic Line Rating solution, leading to increased safety and efficiency in grid operations.
Ingvar Lukas Data Scientist
California’s wind climate is shaped by a combination of large-scale atmospheric circulation patterns and diverse regional topography. From coastal breezes along the Pacific Ocean to desert gusts in the southeast, wind speeds and directions in the Golden State can vary dramatically over short distances.
The interplay of climate systems, topographic features, and land-sea temperature contrasts makes California a great case study for hyperlocal machine learning solutions that improve on large-scale numerical weather prediction models.
General Wind Patterns
Spring-Early Fall
During this period, the North Pacific High (a semi-permanent high-pressure system off the West Coast) dominates California’s weather. Coastal and valley areas experience a regular sea breeze, with mild to moderate afternoon winds.
Late Fall-Winter
As the Pacific High weakens or shifts westward, frontal systems more frequently impact the state, bringing stronger westerly or southwesterly winds, especially in Northern and Central California.
Santa Ana Winds
Often associated with Southern California, these offshore, hot, dry, and gusty winds typically blow in fall and winter. Speeds of 10-15 m/s are common, with gusts exceeding 25 m/s in canyon or pass areas (e.g., Santa Clarita, Malibu).
Influence of Topography and Localized Effects
Channeling & Microclimates
Coastal plains, mesas, and inland valleys create localized channeling effects, where terrain can amplify or diminish wind speeds. Bluffs enhance uplift and turbulence, while canyons funnel winds to locally higher velocities.
Urban Development & Land Cover
Built environments, forests, and other land cover v variations significantly influence near-surface wind flow. These small-scale effects can cause existing weather forecasts to be insufficiently accurate, especially in complex terrain.
As a result, precise wind predictions all over the state are vital for unlocking the full potential of transmission and distribution assets. However, standard numerical weather models often struggle to capture fine-grained variability, prompting the need for hyperlocal machine learning approaches.
Improving accuracy by 50% — San Diego · Sunset Oaks Drive
Below are examples (for both summer and winter cases) from the San Diego region, showcasing wind speed measurements, numerical weather predictions (nwp ), and Gridraven’s machine learning model, with 95% confidence intervals. Summer Winter August 1 – 15, 2024
February 11 – 27, 2024
The model had no prior access to these specific measurements or locations (i.e., they were in the test set). Evidently, the observed winds stay within the 95% confidence intervals.
Find out how much more your grid can really handle right now, and preview our product at claw.gridraven.com. This shows the capacity of the entire world’s grid.
Founded by Power industry professionals
Before launching Gridraven, Georg led digitalisation at Estonia’s national grid, Elering. He spearheaded the digitalisation task force, crafted the digital strategy, and guided IT investments. As head of the Smart Grid Unit, Georg also helped build a Europe-wide energy data platform.
Before that, Georg was the CTO and co-founder of Sympower, which now manages 1.5GW of grid flexibility and a team of 200+. He holds a Master’s in Sustainable Energy from Imperial College London.
Georg’s all about pushing the limits of the grid to power a brighter future.
Henri is an electrical power engineer with a passion for using cuttingedge tech to make power systems smarter and more efficient. In 2022, he defended his PhD, where he applied machine learning to assess the condition of overhead transmission lines. He didn’t stop there—Henri took this innovation to the Estonian grid.
With seven years at Elering, he’s been the brains behind riskbased asset management systems for electricity and gas networks. Now, he’s bringing that expertise to Gridraven while still pushing power system research at Tallinn University of Technology. He’s also helping shape global standards for transmission systems at CIGRE.
Henri is all about pushing the boundaries of what’s possible in the energy sector.
Georg Rute CEO
Henri Manninen, PhD CTO
Markus Lippus Chief Data Scientist
Before co-founding Gridraven, Markus was the Chief Data Scientist and technical founder of MindTitan, an AI powerhouse in Estonia. He has personally built around a hundred AI solutions and was the go-to technical person at MindTitan to solve any issues. He has helped industries and governments turn AI from research into real-world game changers that made everything smarter and more efficient. He has also served as an expert in developing AI strategies for the Estonian government and public sector organisations in KSA.
Markus is all about pushing the limits of how AI can be applied in the real world to transform industries and make everything smarter and more efficient.
BACKGROUND
Electricity demand is growing
Power demand is increasing thanks to electrification, AI and electric vehicles. Meeting rising demand is increasingly challenging, placing the grids under double stress.
Solar and wind are accelerating
The costs of solar photovoltaic modules and wind turbines continue to fall. In many places around the world, solar and wind are now the cheapest form of energy, which is driving an exponential built out of these intermittent power sources.
Additional capacity is not available on day-ahead markets.
Existing Dynamic Line Rating solutions depend on hardware that measure real-time solutions. But at least 36 hour forecasts are needed to make more capacity available for energy markets. The majority of power is bought and sold day-ahead.
Maximise existing grid capacity
What can be done now? The clearest vision comes from the German regulator in their network development plan - the NOVA principle. It means that the grid should first be optimized and only then reinforced or new lines built (Netz-Optimierung vor Verstärkung vor Ausbau). The quickest win is to utilize the existing assets as efficiently as possible.
Weather determines the maximum transmission capacity of overhead lines
The risk of overheating fundamentally limits high-voltage overhead lines. Power flowing through the line heats up the conductors, and each line has a maximum temperature it can tolerate. But when it’s cold or windy, the same power line can carry several times more power before overheating. Dynamic Line Rating (DLR) is a mature technology that can unlock up to 30% more capacity from the existing grid.
WHY DYNAMIC LINE RATINGS
Dynamic Line Ratings unlock up to 30% additional capacity over the year
The most significant benefit comes from the wind-cooling effect. Dynamic line ratings use the wind forecast and often lead to twice the capacity. However, because accurately predicting wind is difficult, dynamic ratings are not yet widely used.
More cheap, clean energy thanks to forecasting line ratings day-ahead
Electricity is bought and sold a day ahead. Turning thermal power plants on and off takes hours, so production schedules are planned in advance either on power markets or by central dispatchers. Only with day-ahead forecasting can dispatchers and markets take advantage of additional grid capacity and allow more cheap power onto the grid.
Gridraven unlocks more capacity for day-ahead energy markets thanks to its ability to confidently forecast line ratings multiple days into the future.
IMPACT
More revenue
Grid operators can safely carry more power through their network, earning more revenue from energy fees. Integrated utilities can sell more power to consumers.
Cheaper energy
More grid capacity means less curtailment and more interconnection, which both tend to reduce the cost of energy for consumers.
Reliable power
Accurate line ratings provide more options for system operators thanks to additional capacity.