High-voltage transmission towers at dusk
Apache 2.0PythonNOAA · NASA · USDA · USFS

Climate intelligence for every utility, not just the ones with big budgets.

climagrid turns public federal weather data into IEEE and ASCE standards-based stress scores, so rural electric cooperatives can prioritize maintenance and strengthen grid resilience before weather does the damage.

Built for the utilities serving 42 million Americans across 56% of the U.S. landmass, with no expensive software or data science team required.

Get Started View on GitHub Documentation
IEEE C57.91 thermal aging ASCE ice & freeze-thaw loading Tested on real grid events
The problem

Small utilities carry big grid risk, with the fewest tools to manage it.

Rural electric cooperatives and small municipal utilities maintain a vast share of America's grid, yet rarely have the budget or staff for the predictive maintenance systems that large investor-owned utilities rely on every day.

Predictive tools are priced out of reach

Commercial asset-health platforms cost six figures a year. A co-op serving a few thousand meters across hundreds of miles of line is left inspecting on fixed schedules instead of prioritizing the assets weather is actually punishing.

Weather stress stays invisible until something fails

Heat waves silently age transformers, ice loads overstress conductors, and freeze-thaw cycles weaken poles long before anything visibly breaks. Without data, small crews are reacting to outages rather than preventing them.

What it measures

Engineering standards, applied to your assets automatically.

Each stress model implements a published, peer-reviewed standard. No black boxes and no proprietary scoring, just transparent calculations you can audit and trust.

IEEE C57.91

Transformer thermal aging

Computes hot-spot temperature and loss-of-life using the exact IEEE C57.91 loading guide, so you know which transformers heat waves are aging fastest.

  • Hot-spot & top-oil modeling
  • Per-unit loss-of-life
  • Heat-wave accumulation
ASCE 7 / NESC

Ice loading & freeze-thaw

Applies ASCE ice-accretion and combined wind-on-ice loading to lines and poles, and tracks freeze-thaw cycles that fatigue structures over time.

  • Radial ice accretion
  • Wind-on-ice combined loads
  • Freeze-thaw cycle counts
USFS / NASA

Wildfire risk exposure

Blends fuel, drought, and fire-weather indices from USFS and NASA sources to flag spans and poles in elevated wildfire-stress conditions.

  • Fire-weather indexing
  • Drought & fuel signals
  • Span-level exposure scoring
Trust & validation

Built to be trusted by the engineers who run the grid.

climagrid is transparent by design. There is nothing proprietary to take on faith, only documented standards, reproducible math, and open code.

Exact published standards

Calculations follow IEEE C57.91, ASCE 7, and NESC line-loading guidance to the letter, using the same references utility engineers already rely on.

Validated on real events

Stress models were tested against documented extreme weather, including the 2021 Pacific Northwest heat dome, to confirm they track real-world equipment strain.

Fully open source

Every model, data source, and assumption is public under the Apache 2.0 license. Inspect it, extend it, and verify it against your own field experience.

Cite the release · DOI 10.5281/zenodo.20256536
The impact

Tools that match the scale of rural America's grid.

The utilities that maintain the most line per customer have historically had the least access to predictive analytics. climagrid is built to change that balance.

0+

Rural electric cooperatives that could benefit

0M

Americans served by these utilities

0%

Of U.S. land area covered by co-op service territory

Leadership

The people behind climagrid.

A team blending rigorous ML research with proven energy infrastructure execution, united by one goal: making grid resilience accessible to every utility.

Temidire Adesiji, Technical Founder at climagrid

Temidire Adesiji

Technical Founder

Temidire is the technical founder of climagrid and an empirical ML systems researcher at Brown University, focused on building reliable AI systems.

He has driven measurable impact, fine-tuning CNNs for predictive maintenance at Chevron and cutting equipment failure rates by 30%. As co-founder of MingoolAI, he built an autonomous AI Chief of Staff using LLM orchestration and multi-agent workflows.

A McKinsey Forward Program Fellow and former National Mathematics Olympiad Gold Medalist, he combines rigorous research with hands-on systems engineering to strengthen grid resilience for rural cooperatives through open climate data.

Tomiwa Adebayo, Head of Strategy, Partnerships & Growth at climagrid

Tomiwa Adebayo

Head of Strategy, Partnerships & Growth

Tomiwa leads strategy, partnerships, and growth at climagrid, helping rural electric cooperatives turn public federal climate data into tools for grid resilience, predictive maintenance, and the clean energy transition.

As Founder and CEO of Carima, he built and scaled one of West Africa's early EV charging networks, deploying over 100 sites and serving more than 25,000 customers.

A UC Berkeley Haas MBA candidate in Energy and Mobility with experience at Amazon and Allianz, he brings proven infrastructure execution together with deep strategic and institutional insight.

Try it

See a stress score take shape.

A simplified, illustrative preview of how weather inputs map to a relative stress score. The actual toolkit uses full published standards and real public data feeds.

Equipment type

38°C
85%
10mm

Relative weather stress score

62/ 100
Elevated stressPrioritize inspection

Illustrative only. Scores prioritize where to look first. They do not predict failure.

Get started

Up and running in one command.

climagrid is a pure-Python package on PyPI. Install it, point it at your service area, and start scoring weather stress against public data feeds.

terminal
$ pip install climagrid
Read the docs GitHub repository PyPI package
Tech foundation & roadmap

A solid engineering foundation, built to grow.

climagrid stands on published standards and public data today, and is designed from the ground up to feed the predictive models utilities will build tomorrow.

Weather data
Standards
Features
Future ML

Current foundation

Public-data pipeline

Built in Python on open data from NOAA, NASA, USDA, and USFS, so every input is traceable to a public source.

Published standards

Implements documented engineering standards, including IEEE C57.91 for transformer thermal aging and ASCE 7-22 for ice loading.

ML-ready features

Produces clean, explainable, ML-ready features for each piece of equipment, ready for analysis or modeling.

Future direction

Available now

Explainable feature outputs

Output features are deliberately designed to serve as high-quality inputs for machine learning models, not opaque scores.

Next

Predictive maintenance models

Integrate with predictive maintenance and failure-prediction models once utilities contribute historical equipment failure data.

Long term

Climate scenarios & prioritization

Support climate scenario modeling and automated, risk-based maintenance prioritization across a full service territory.

From the field

The engineering behind the scores.

Plain-language writing on the standards, weather signals, and field operations that power resilient rural grids. Written for the engineers who keep the lights on.

Read the blog
Get in touch

Request a demo or share feedback.

Whether you run a co-op, a municipal utility, or you're a researcher building on climagrid, we'd love to hear how you want to use it.

  • Utilities · request a guided demo on your service territory
  • Engineers · suggest a standard or data source to support next
  • Contributors · open an issue or PR on GitHub anytime