A Review of AI Applications in Unconventional Oil and Gas Exploration and Development

Energies, Год журнала: 2025, Номер 18(2), С. 391 - 391

Опубликована: Янв. 17, 2025

The development of unconventional oil and gas resources is becoming increasingly challenging, with artificial intelligence (AI) emerging as a key technology driving technological advancement industrial upgrading in this field. This paper systematically reviews the current applications trends AI exploration development, covering major research achievements geological exploration; reservoir engineering; production forecasting; hydraulic fracturing; enhanced recovery; health, safety, environment management. how deep learning helps predict distribution classify rock types. It also explains machine improves simulation history matching. Additionally, we discuss use LSTM DNN models forecasting, showing has progressed from early experiments to fully integrated solutions. However, challenges such data quality, model generalization, interpretability remain significant. Based on existing work, proposes following future directions: establishing standardized sharing labeling systems; integrating domain knowledge engineering mechanisms; advancing interpretable modeling transfer techniques. With next-generation intelligent systems, will further improve efficiency sustainability development.

Язык: Английский

---

Predictive modeling of reservoir geomechanical parameters through computational intelligence approach, integrating core and well logging data

Earth Science Informatics, Год журнала: 2025, Номер 18(1)

Опубликована: Янв. 1, 2025

Язык: Английский

---

A Novel Hybrid Bayesian-Group-Based Machine Learning (HB-GML) Method for Predicting Uniaxial Compressive Strength (UCS) of Rock

Rock Mechanics and Rock Engineering, Год журнала: 2025, Номер unknown

Опубликована: Янв. 20, 2025

Язык: Английский

---

Regression models to estimate total organic carbon (TOC) from well-log data

Elsevier eBooks, Год журнала: 2025, Номер unknown, С. 43 - 77

Опубликована: Янв. 1, 2025

Язык: Английский

---

Comparison of machine learning models for rock UCS prediction using measurement while drilling data

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 11, 2025

Accurately determining the uniaxial compressive strength (UCS) of rocks is crucial for various rock engineering applications. However, traditional methods obtaining UCS are often time-consuming, labor-intensive, and unsuitable fractured sections. In recent years, using Measurement-while-drilling data to identify has gained traction as an alternative approach. To develop a method that can rapidly, efficiently, economically estimate across different types conditions based on while-drilling tests, this study compiles comprehensive dataset from existing literature. The includes drilling parameters their corresponding values, collected under varying lithologies, levels, drill bit types, conditions. Five machine learning models—multilayer perceptron (MLP), support vector regression (SVR), convolutional neural networks (CNN), random trees (RT), long short-term memory (LSTM)—were trained evaluated. Among these, RT demonstrated superior predictive performance, achieving root mean square error (RMSE) 15.851, absolute (MAE) 4.449, standard deviation residuals (SDR) 15.292, R² value 0.959 test set. SVR also performed well, with RMSE 21.905, MAE 17.962, SDR 21.144, 0.922. While CNN LSTM exhibited slightly higher errors, they showed better generalization capabilities validation datasets. Furthermore, models were validated unseen independent dataset, where achieved best results, followed by SVR, while other relatively poorly. This indicates demonstrate suitability prediction.

Язык: Английский

---

Robust Machine Learning Predictive Models for Real-Time Determination of Confined Compressive Strength of Rock Using Mudlogging Data

Rock Mechanics and Rock Engineering, Год журнала: 2024, Номер 57(9), С. 6881 - 6907

Опубликована: Апрель 11, 2024

Язык: Английский

---

Advanced machine learning approaches for uniaxial compressive strength prediction of Indian rocks using petrographic properties

Multiscale and Multidisciplinary Modeling Experiments and Design, Год журнала: 2024, Номер 7(6), С. 5265 - 5286

Опубликована: Июль 3, 2024

Язык: Английский

---

Detection methods for strength deterioration and structural characteristics of fractured rock based on digital drilling

Measurement, Год журнала: 2024, Номер 233, С. 114779 - 114779

Опубликована: Апрель 25, 2024

Язык: Английский

---

A smarter approach to liquefaction risk: harnessing dynamic cone penetration test data and machine learning for safer infrastructure

Frontiers in Built Environment, Год журнала: 2024, Номер 10

Опубликована: Окт. 25, 2024

This paper presents a novel approach for assessing liquefaction potential by integrating Dynamic Cone Penetration Test (DCPT) data with advanced machine learning (ML) techniques. DCPT offers cost-effective, rapid, and adaptable method evaluating soil resistance, making it suitable assessment across diverse conditions. study establishes threshold criterion based on the ratio of penetration rate to dynamic resistance ( e / q d ), where values exceeding four indicate high susceptibility. ML models, including Support Vector Machine (SVM) optimized Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), Genetic Algorithm (GA), Firefly (FA), were employed predict using key geotechnical parameters, such as fine content, peak ground acceleration, reduction factor, rate. The SVM-PSO model demonstrated superior performance, R 2 0.999 0.989 in training testing phases, respectively. proposed methodology sustainable accurate assessment, reducing environmental impact investigations, while ensuring reliable predictions. bridges gap between field computational techniques, providing powerful tool engineers assess risks design resilient infrastructures.

Язык: Английский

---

Combined Deep Learning and Optimization for Hydrogen-Solubility Prediction in Aqueous Systems Appropriate for Underground Hydrogen Storage Reservoirs

Energy & Fuels, Год журнала: 2024, Номер 38(22), С. 22031 - 22049

Опубликована: Окт. 30, 2024

The widespread use of fossil fuels drives greenhouse gas emissions, prompting the need for cleaner energy alternatives like hydrogen. Underground hydrogen storage (UHS) is a promising solution, but measureing (H2) solubility in brine complex and costly. Machine learning can provide accurate reliable predictions H2 by analyzing diverse input variables, surpassing traditional methods. This advancement crucial improving UHS, making it viable component sustainable infrastructure. Given its importance, this study utilized convolutional neural network (CNN) long–short-term memory (LSTM) deep algorithms combination with growth optimization (GO) gray wolf (GWO) to predict solubility. A total 1078 data points were collected from laboratory results, including variables temperature (T), pressure (P), salinity (S), salt type (ST). After removing 97 points, which identified as outliers duplicates, remaining 981 divided into training testing sets using best separation ratio selected based on sensitivity analysis. Standalone hybrid forms then applied develop predictive models optimized control parameters both algorithms. Among developed models, CNN-GO has lowest root-mean-square error (RMSE, train: 0.00006 mole fraction test: 0.00021 fraction) compared other standalone models. application scoring regression characteristic (REC) curve analysis showed that model generated prediction performance. Shapley additive explanation indicated P was most important factor influencing solubility, followed S, T, ST, order. Partial dependency revealed ability capture nonlinear relationships between features target variable.

Язык: Английский

---