Speaker
Description
The intricate nature of subsurface carbon mineralization reactive transport processes has driven researchers to employ artificial intelligence (AI) and machine learning (ML) for analyzing extensive experimental and monitoring and simulation-based data collections. The research assesses present AI-based investigation methods which use data-driven approaches to study carbon mineralization and reactive transport in porous and fractured systems for better physical modeling.
The review presents an overview of ML applications which analyze geochemical data and calculate reaction parameters and simulate complex reactive transport processes. The research evaluates ML-based surrogate models which replace reaction solvers and perform sensitivity analysis and uncertainty exploration through their reported boundaries and their identified constraints. The research employs data-driven methods to detect mineral dissolution and precipitation patterns which emerge from simulated and experimental data.
The review examines the application of physics-informed ML methods which merge mass conservation and reaction kinetics equations into learning models for reactive transport modeling. Existing studies are discussed in terms of their reported ability to improve physical consistency and stability relative to purely data-driven models. The literature shows three main challenges which include stiff reaction kinetics and scale dependency and the limited availability of sparse data.
The paper examines AI applications for monitoring systems and data integration through its evaluation of pattern recognition methods for chemical concentration time series and mineralization process spatial data. The review explains how AI systems handle different types of data while helping users understand complicated reactive systems yet it identifies multiple issues which affect system interpretation and data transferability and uncertainty measurement.
The review unites previous studies to demonstrate how AI functions as an analytical instrument which improves physics-based porous media modeling systems for carbon mineralization research.
| Country | India |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
