Carbon capture, utilization and sequestration systems design and operation optimization: Assessment and perspectives of artificial intelligence opportunities

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 917, P. 170085 - 170085

Published: Jan. 15, 2024

Language: Английский

---

Catalyzing net-zero carbon strategies: Enhancing CO2 flux Prediction from underground coal fires using optimized machine learning models

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 441, P. 141043 - 141043

Published: Jan. 31, 2024

Language: Английский

---

A comparative study of statistical and machine learning models on carbon dioxide emissions prediction of China

Environmental Science and Pollution Research, Journal Year: 2023, Volume and Issue: 30(55), P. 117485 - 117502

Published: Oct. 23, 2023

Language: Английский

---

Enhancing carbon sequestration: Innovative models for wettability dynamics in CO2-brine-mineral systems

Journal of environmental chemical engineering, Journal Year: 2024, Volume and Issue: 12(5), P. 113435 - 113435

Published: June 26, 2024

Language: Английский

---

A Comprehensive Summary of the Application of Machine Learning Techniques for CO2-Enhanced Oil Recovery Projects

Machine Learning and Knowledge Extraction, Journal Year: 2024, Volume and Issue: 6(2), P. 917 - 943

Published: April 29, 2024

This paper focuses on the current application of machine learning (ML) in enhanced oil recovery (EOR) through CO2 injection, which exhibits promising economic and environmental benefits for climate-change mitigation strategies. Our comprehensive review explores diverse use cases ML techniques CO2-EOR, including aspects such as minimum miscible pressure (MMP) prediction, well location optimization, production factor multi-objective Pressure–Volume–Temperature (PVT) property estimation, Water Alternating Gas (WAG) analysis, CO2-foam EOR, from 101 reviewed papers. We catalog relative information, input parameters, objectives, data sources, train/test/validate results, evaluation, rating score each area based criteria quality, ML-building process, analysis results. also briefly summarized limitations methods petroleum industry applications. detailed extensive study could serve an invaluable reference employing industry. Based review, we found that offer great potential solving problems majority CO2-EOR areas involving prediction regression. With generation massive amounts everyday gas industry, can provide efficient reliable preliminary results

Language: Английский

---

Sub-surface geospatial intelligence in carbon capture, utilization and storage: A machine learning approach for offshore storage site selection

Energy, Journal Year: 2024, Volume and Issue: 305, P. 132086 - 132086

Published: June 19, 2024

This study introduces an innovative data-driven and machine-learning framework designed to accurately predict site scores in the screening for specific offshore CO2 storage sites. The seamlessly integrates diverse sub-surface geospatial data sources with human aided expert-weighted criteria, thereby providing a high-resolution tool. Tailored accommodate varying accessibility significance of this approach considers both technical non-technical factors. Its purpose is facilitate identification priority locations projects associated Carbon Capture, Utilization, Storage (CCUS). Through aggregating analyzing datasets, employs machine learning algorithms model identify suitable geologic CCUS regions. process adheres stringent safety, risk control, environmental guidelines, addressing situations where analysis may fail recognize patterns provide detailed insights techniques. primary emphasis research bridge gap between scientific inquiry practical application, facilitating informed decision-making implementation projects. Rigorous assessments encompassing geological, oceanographic, eco-sensitivity metrics contribute valuable policymakers industry leaders. To ensure accuracy, efficiency, scalability established facilities, proposed undergoes benchmarking. comprehensive evaluation includes utilization such as Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multilayer Learning Machine (MLELM), Deep Neural Network (DNN) predicting more scores. Among these algorithms, DNN algorithm emerges most effective score prediction. strengths encompass nonlinear modeling, feature learning, scale invariance, handling high-dimensional data, end-to-end transfer representation parallel processing. results demonstrate high accuracy testing subset, values AAPD (Average Absolute Percentage Difference) = 1.486%, WAAPD (Weighted Average 0.0149%, VAF (Variance Accounted For) 0.9937, RMSE (Root Mean Square Error) 0.9279, RSR Sum Squares Residuals) 0.0068, R2 (Coefficient Determination) 0.9937.

Language: Английский

---

Enhanced hydrogen sequestration through hydrogen foam generation with SDS Co-Injection in subsurface formations

Chemical Engineering Journal, Journal Year: 2024, Volume and Issue: 481, P. 148575 - 148575

Published: Jan. 6, 2024

Language: Английский

---

An application of a genetic algorithm in co-optimization of geological CO2 storage based on artificial neural networks

Clean Energy, Journal Year: 2024, Volume and Issue: 8(1), P. 111 - 125

Published: Jan. 10, 2024

Abstract Global warming, driven by human-induced disruptions to the natural carbon dioxide (CO2) cycle, is a pressing concern. To mitigate this, capture and storage has emerged as key strategy that enables continued use of fossil fuels while transitioning cleaner energy sources. Deep saline aquifers are particular interest due their substantial CO2 potential, often located near fuel reservoirs. In this study, deep aquifer model with water production well was constructed develop optimization workflow. Due time-consuming nature each realization numerical simulation, we introduce surrogate derived from extracted data. The novelty our work lies in pioneering simultaneous using machine learning within an integrated framework. Unlike previous studies, which typically focused on single-parameter optimization, research addresses gap performing multi-objective for breakthrough time data-driven model. Our methodology encompasses preprocessing feature selection, identifying eight pivotal parameters. Evaluation metrics include root mean square error (RMSE), absolute percentage (MAPE) R2. predicting values, RMSE, MAPE R2 test data were 2.07%, 1.52% 0.99, respectively, blind data, they 2.5%, 2.05% 0.99. For time, 2.1%, 1.77% 0.93, 2.8%, 2.23% 0.92, respectively. addressing computational demands coupling simulator algorithm, have adopted trained artificial neural network seamlessly genetic algorithm. Within framework, conducted 5000 comprehensive experiments rigorously validate development Pareto front, highlighting depth approach. findings study promise insights into interplay between aquifer-based processes framework based coupled optimization.

Language: Английский

---

Low-Carbon Advancement through Cleaner Production: A Machine Learning Approach for Enhanced Hydrogen Storage Predictions in Coal Seams

Renewable Energy, Journal Year: 2025, Volume and Issue: unknown, P. 122342 - 122342

Published: Jan. 1, 2025

Language: Английский

---

Nanoparticles in enhanced oil recovery: state-of-the-art review

Journal of Petroleum Exploration and Production Technology, Journal Year: 2025, Volume and Issue: 15(4)

Published: March 12, 2025

Language: Английский

---