Request for Information (RFI)- Machine Learning for Geothermal Energy and the Geosciences | DE FOA 0001920

Opportunity Information: Apply for DE FOA 0001920

The Department of Energy (DOE), through the Office of Energy Efficiency and Renewable Energy (EERE) and its Geothermal Technologies Office (GTO), issued this Request for Information (RFI) to collect public feedback on how machine learning (ML) could be applied to geoscience problems that matter for geothermal energy. The central idea is that geothermal development depends heavily on interpreting complex subsurface data, and DOE wants to understand where modern ML methods could most improve research, development, and decision-making. A key theme running through the RFI is that geothermal and the broader geosciences already generate rich datasets, but their value is not fully realized without better tools, shared standards, and community infrastructure that can support advanced ML workflows.

The RFI is specifically organized around three input areas. First, DOE is asking stakeholders to identify the most promising ML applications for subsurface research and development (R&D) relevant to geothermal energy. This includes use cases where ML could accelerate or improve tasks like subsurface characterization, interpretation of geological and geophysical signals, resource assessment, reservoir modeling, exploration targeting, and risk reduction in drilling and field development. The goal is to surface the highest-impact opportunities where ML could change the pace or cost of geothermal innovation, especially in parts of the workflow where uncertainty is high, data are heterogeneous, and traditional methods can be slow or expensive.

Second, DOE is seeking ideas on building open, community datasets that are suitable for state-of-the-art ML. That emphasis on "open community datasets" reflects a recognition that many of the best ML breakthroughs depend on accessible, well-curated training data, benchmarks, and shared evaluation metrics. DOE is looking for guidance on what datasets would be most valuable to create or release, how they should be formatted, what metadata and labeling standards are needed, and what practical barriers exist (such as proprietary constraints, inconsistent collection methods, limited ground truth, or the need for anonymization). The underlying intent is to enable reproducible ML research in geothermal and geoscience settings, where data are often scattered across organizations and stored in non-standard forms.

Third, the RFI asks for feedback on how to leverage crowd-sourced R&D through alternative funding mechanisms. This points to approaches beyond conventional single-team research awards, such as prize challenges, open innovation competitions, cooperative benchmarking efforts, or other models that can mobilize a wider technical community. By exploring these mechanisms, DOE is signaling interest in attracting contributions from data scientists and ML practitioners who may not typically work in geothermal, while also creating ways to test solutions rapidly and compare methods fairly using shared datasets and evaluation criteria.

A notable aspect of this RFI is that DOE explicitly welcomes input from outside the geothermal community. Because ML is more mature in other sectors (for example, technology, finance, autonomous systems, and parts of oil and gas), DOE is inviting perspectives from any industry that works with subsurface data or faces similar challenges such as sparse measurements, noisy signals, complex physical processes, and high-cost field operations. The RFI also flags potential crosscutting value: methods developed for geothermal could translate to other subsurface industries, and partnerships across sectors are of interest, especially where data, tools, and expertise can be shared.

Operationally, this announcement is not a Funding Opportunity Announcement (FOA) and does not accept project applications. It is a market and stakeholder feedback exercise meant to inform DOE decision-making before DOE potentially releases a future FOA. Responses were required to be submitted by email as a Microsoft Word attachment to machinelearninggeo@ee.doe.gov, with a deadline of 5:00 PM EDT on June 6, 2018. The opportunity appeared on EERE Exchange under Funding Opportunity Number DE-FOA-0001920, categorized as discretionary and energy-related, with eligibility listed as unrestricted (open to any entity type, subject to any additional eligibility details in the full text). The administering office is DOE's Golden Field Office, and the official posting date was May 7, 2018.

• The Department of Energy, Golden Field Office in the energy sector is offering a public funding opportunity titled "Request for Information (RFI)- Machine Learning for Geothermal Energy and the Geosciences" and is now available to receive applicants.
• Interested and eligible applicants and submit their applications by referencing the CFDA number(s): 81.087.
• This funding opportunity was created on May 07, 2018.
• Applicants must submit their applications by Jun 06, 2018 All responses to this RFI must be provided as an attachment (in Microsoft Word format) to an e-mail message addressed to machinelearninggeo@ee.doe.gov.. (Agency may still review applications by suitable applicants for the remaining/unused allocated funding in 2026.)
• Each selected applicant is eligible to receive up to $2.00 in funding.
• The number of recipients for this funding is limited to 1 candidate(s).
• Eligible applicants include: Unrestricted (i.e., open to any type of entity above), subject to any clarification in text field entitled Additional Information on Eligibility.

Apply for DE FOA 0001920

[Watch] Creating a grant proposal using the step-by-step wizard inside the applicant portal: