Machine learning for sustainable geoenergy

2026-05-05 · Source: A Geodyssey – Geoscience Text Analytics and Enterprise Search Research · Field: Energy & Utilities — Renewable Energy Systems, Emerging Energy Technologies, Energy Efficiency & Conservation · Depth: Expert, quick

Summary

A paper by Menke et al. (2026) from Heriot-Watt University addresses the application of machine learning (ML) in sustainable geoenergy projects, including CO2 storage, geothermal, and subsurface H2 generation. These projects follow a petroleum-style funnel from screening to operations, requiring risk-bounded operational choices despite limited and heterogeneous observations. The authors identify four bottlenecks: scarce data and field outcomes, inadequate uncertainty quantification, weak scale-bridging in physical models, and insufficient quality assurance for regulatory deployment. They propose ML approaches such as hybrid physics-ML, probabilistic uncertainty quantification, and multi-fidelity learning to tackle these issues. The paper connects these methods to applications like digital twins, multiphase flow, monitoring, and basin-scale portfolio management, concluding with an agenda for benchmarks and reporting standards.

Key takeaway

For geoenergy engineers and project managers developing sustainable subsurface projects, you should prioritize ML approaches that explicitly integrate physics and quantify uncertainty. Focusing on hybrid physics-ML and probabilistic uncertainty quantification will improve the credibility of climate-mitigation claims and support regulatory compliance, ultimately controlling deployment rates and capital costs. Consider adopting proposed benchmarks and reporting standards to ensure defensible ML applications.

Key insights

Machine learning can enhance sustainable geoenergy projects by addressing data scarcity and uncertainty through physics-informed methods.

Principles

• Uncertainty quantification is a primary deliverable.
• Integrate physics with ML for robust models.
• Ensure auditability for regulator-facing deployment.

Method

The proposed ML approach combines hybrid physics-ML, probabilistic uncertainty quantification, structure-aware representations, and multi-fidelity/continual learning to address data scarcity and complex physical constraints in geoenergy.

In practice

• Develop imaging-to-process digital twins.
• Improve multiphase flow and near-well conformance.
• Enhance monitoring, measurement, and verification (MMV).

Topics

• Sustainable Geoenergy
• Machine Learning
• Uncertainty Quantification
• CO2 Storage
• Physics-informed ML

Best for: AI Scientist, Research Scientist, Domain Expert

Related on AIssential

• Engineering Hybrid Physics-Informed Neural Networks for Next-Generation Electricity Systems: A State-of-the-Art Review — Physics-informed machine learning integrates domain physics into models, enhancing accuracy and efficiency for data-scarce electricity systems.
• Machine-Learning Emulation of Satellite Greenhouse Gas Retrievals: Stability over Time — Machine-learning emulators for GHG retrievals require time-aware training to maintain accuracy and stability over extended periods.
• A green solvent screening tool for emerging materials via uncertainty aware, transformer enhanced transfer learning — Transfer learning with uncertainty quantification enables accurate green solvent screening despite limited data.
• Porosity and permeability prediction from petrographic point-counting data using machine learning — Machine learning can predict reservoir properties from historical petrographic point-counting data.
• Review of Machine Learning Models for Solar Energetic Particle Prediction — Machine learning offers a promising new approach to predict hazardous Solar Energetic Particle events, complementing traditional methods.
• Uncertainty Quantification of Engineering Structures by Polynomial Chaos Expansion and Multivariate Active Learning — Adaptive sequential sampling for multi-output PCE improves surrogate accuracy and stability by balancing variance exploitation and spatial exploration.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by A Geodyssey – Geoscience Text Analytics and Enterprise Search Research.