SourceGeekWire•Stalecollected in 35m
PNNL, Nvidia, and Fervo Partner on Geothermal AI
#geothermal#digital-twin#clean-energy#simulationgeothermal-digital-twinnvidiapnnlfervo energy
See how Nvidia's digital twin tech is being applied to solve complex geothermal energy challenges.
30-Second TL;DR
What Changed
Partnership between PNNL, Nvidia, and Fervo Energy to optimize geothermal drilling.
Why It Matters
This initiative could significantly lower the cost and risk of geothermal energy projects, making it a more viable competitor to fossil fuels. It demonstrates the growing role of AI in physical infrastructure and resource management.
What To Do Next
Monitor the PNNL and Nvidia research portals for the release of the digital twin tool to explore how subsurface modeling can be applied to your own physical simulation projects.
Who should care:Researchers & Academics
Key Points
- •Partnership between PNNL, Nvidia, and Fervo Energy to optimize geothermal drilling.
- •Development of a public digital twin tool to model subsurface geological conditions.
- •Goal to accelerate clean energy production by eliminating guesswork in deep-earth drilling.
Deep Insight
AI-generated analysis for this event — not the original article.
Enhanced Key Takeaways
- •The project utilizes the NVIDIA Earth-2 climate digital twin platform to integrate high-fidelity subsurface data with AI-driven predictive modeling.
- •PNNL is contributing its specialized subsurface modeling software, known as Subsurface Transport Over Multiple Phases (STOMP), to enhance the accuracy of the digital twin.
- •Fervo Energy is providing proprietary data from its operational geothermal sites, such as Project Red in Nevada, to train and validate the AI models.
- •The collaboration is supported by the U.S. Department of Energy's Geothermal Technologies Office, aligning with national goals to reach 90 GW of geothermal capacity by 2050.
- •The digital twin aims to specifically address the 'blind drilling' problem, where operators face high costs and risks due to the inability to accurately characterize rock permeability and thermal properties before drilling.
Competitor Analysis
Data Integration
- Fervo/PNNL/Nvidia Project
- Real-time digital twin
- Traditional Geothermal Exploration
- Static geological surveys
- Competitor AI Solutions (e.g., Google/Sage Geosystems)
- Limited cloud-based modeling
Drilling Risk
- Fervo/PNNL/Nvidia Project
- AI-predicted subsurface mapping
- Traditional Geothermal Exploration
- High (Trial and error)
- Competitor AI Solutions (e.g., Google/Sage Geosystems)
- Moderate (Predictive analytics)
Compute Power
- Fervo/PNNL/Nvidia Project
- Accelerated GPU-based simulation
- Traditional Geothermal Exploration
- CPU-based legacy systems
- Competitor AI Solutions (e.g., Google/Sage Geosystems)
- Variable cloud compute
| Feature | Fervo/PNNL/Nvidia Project | Traditional Geothermal Exploration | Competitor AI Solutions (e.g., Google/Sage Geosystems) |
|---|---|---|---|
| Data Integration | Real-time digital twin | Static geological surveys | Limited cloud-based modeling |
| Drilling Risk | AI-predicted subsurface mapping | High (Trial and error) | Moderate (Predictive analytics) |
| Compute Power | Accelerated GPU-based simulation | CPU-based legacy systems | Variable cloud compute |
Technical Deep Dive
- Utilizes NVIDIA Modulus, a framework for developing physics-informed machine learning models, to solve partial differential equations governing subsurface fluid flow.
- Integrates PNNL's STOMP simulator to provide ground-truth physics constraints, ensuring the AI model adheres to thermodynamic and hydrological laws.
- Employs GPU-accelerated seismic inversion techniques to process large-scale geological datasets significantly faster than traditional CPU-based clusters.
- Implements a multi-fidelity modeling approach that combines sparse field measurements with dense synthetic data generated by high-resolution simulations.
Future ImplicationsAI analysis grounded in cited sources
Geothermal drilling success rates will increase by at least 20% within three years.
The reduction of subsurface uncertainty through digital twin modeling directly correlates to fewer dry holes and more precise well placement.
The platform will become an open-source standard for geothermal developers.
By involving a national laboratory like PNNL, the project is structured to disseminate findings to the broader industry to accelerate clean energy adoption.
Timeline
2021-11
Fervo Energy completes the first successful horizontal well test for geothermal energy.
2023-07
Fervo Energy begins operations at Project Red, the world's first commercial enhanced geothermal system.
2024-03
NVIDIA announces the Earth-2 platform for climate and weather digital twins.
2025-09
PNNL and industry partners initiate the integration of AI workflows into subsurface energy research.
2026-05
Formal announcement of the tripartite partnership between PNNL, Nvidia, and Fervo.
- 2021-11Fervo Energy completes the first successful horizontal well test for geothermal energy.
- 2023-07Fervo Energy begins operations at Project Red, the world's first commercial enhanced geothermal system.
- 2024-03NVIDIA announces the Earth-2 platform for climate and weather digital twins.
- 2025-09PNNL and industry partners initiate the integration of AI workflows into subsurface energy research.
- 2026-05Formal announcement of the tripartite partnership between PNNL, Nvidia, and Fervo.
Weekly AI Recap
Read this week's curated digest of top AI events →
Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: GeekWire ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
The weekly digest
One email a week. Unsubscribe anytime.