Porosity prediction of tight reservoir rock using well logging data and machine learning DOI Creative Commons

Yawen He,

Hongjun Zhang,

Zhiyu Wu

и другие.

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

Опубликована: Апрель 16, 2025

Язык: Английский

A novel hybrid machine learning and explainable artificial intelligence approaches for improved source rock prediction and hydrocarbon potential in the Mandawa Basin, SE Tanzania DOI
Christopher N. Mkono, Chuanbo Shen,

Alvin K. Mulashani

и другие.

International Journal of Coal Geology, Год журнала: 2025, Номер unknown, С. 104699 - 104699

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

2

A novel hybrid group method of data handling and Levenberg Marquardt model for estimating total organic carbon in source rocks with explainable artificial intelligence DOI
Christopher N. Mkono, Chuanbo Shen,

Alvin K. Mulashani

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 144, С. 110137 - 110137

Опубликована: Янв. 27, 2025

Язык: Английский

Процитировано

1

Improved Reservoir Porosity Estimation Using an Enhanced Group Method of Data Handling with Differential Evolution Model and Explainable Artificial Intelligence DOI
Christopher N. Mkono, Chuanbo Shen,

Alvin K. Mulashani

и другие.

SPE Journal, Год журнала: 2025, Номер unknown, С. 1 - 19

Опубликована: Фев. 1, 2025

Summary Reservoir characterization is critical to the oil and gas industry, influencing field development, production optimization, hydraulic fracturing, reserves estimation decisions. Accurately estimating porosity crucial for reservoir characterization, well planning, optimization in industry. Traditional determination methods, such as porosimetry, geostatistical, core analysis, often involve complex geological geophysical models, which are expensive time-consuming. This study used integrated machine learning model of differential evolution (DE) with group method data handling (GMDH-DE) estimate using log from Mpyo field, Uganda. The GMDH-DE demonstrates superior performance compared conventional GMDH, support vector regression (SVR), random forest (RF), achieving a coefficient (R2) 0.9925 root mean square error (RMSE) 0.0017 during training, an R² 0.9845 RMSE 0.0121 testing, when validated R2 was 0.9825 0.00018. A key novelty this work integration Shapley additive explanations (SHAP), provides interpretable analysis model’s input features. SHAP reveals that bulk density (RHOB) neutron (NPHI) most parameters estimation, offering valuable insight into features importance. proposed represent novel independent approach accurate interpretability, significantly enhancing efficiency reliability hydrocarbon exploration development.

Язык: Английский

Процитировано

1

Model Development for Brittleness Index Estimation and Depth Determination in Hydraulic Fracturing Operations in Shale Gas Reservoirs Using Machine Learning Algorithms DOI
Grant Charles Mwakipunda,

Norga Alloyce Komba,

Edwin Twum Ayimadu

и другие.

SPE Journal, Год журнала: 2025, Номер unknown, С. 1 - 22

Опубликована: Май 1, 2025

Summary Accurate estimation of the brittleness index (BI) is critical for optimizing hydraulic fracturing operations in shale gas reservoirs, as it directly influences fracture propagation and recovery efficiency. The BI quantifies resistance rock to fracturing, a key factor determining optimal depth stimulation. Prior methods estimating BI, such empirical correlations other utilized machine learning (ML) techniques, often suffer from limited accuracy generalizability, particularly complex geological formations like Fuling field. To address these limitations, ML techniques have gained prominence due their ability capture complex, nonlinear relationships within large data sets, improving predictive accuracy. In this study, we propose novel approach that utilizes hybrid group method handling based on discrete differential evolution (GMDH-DDE) predict BI. GMDH-DDE model was compared with (GMDH), random forest (RF), multilayer perceptron (MLP). results demonstrate significantly outperforms models, achieving coefficient determination (R2) 0.9984, root mean square error (RMSE) 0.2895, absolute (MAE) 0.02543 unseen data. GMDH ranked second estimation, an R2 0.9805, RMSE 0.4635, MAE 0.04224. It followed by RF model, 0.9599, 0.6034, 0.0997. MLP however, had lowest performance, 0.9263, 0.9566, 0.1256. Additionally, demonstrates superior computational efficiency, requiring only 1.12 seconds. This significant advantage over methods, taking 4.82 seconds, 11.23 27.45 These findings highlight potential providing accurate computationally efficient estimations. improved efficiency are expected contribute more effective cost-efficient operations, ultimately enhancing economic viability reservoirs.

Язык: Английский

Процитировано

0

Permeability Prediction and Potential Site Assessment for CO2 Storage from Core Data and Well-Log Data in Malay Basin Using Advanced Machine Learning Algorithms DOI Creative Commons

Md. Yeasin Arafath,

AKM Eahsanul Haque,

Numair Ahmed Siddiqui

и другие.

ACS Omega, Год журнала: 2025, Номер 10(6), С. 5430 - 5448

Опубликована: Фев. 5, 2025

Establishing a potential site characterization for carbon dioxide (CO2) storage in geological formations anticipates the appropriate reservoir properties, such as porosity, permeability, and so forth. Well logs seismic data were utilized to determine key including volume of shale, water saturation. These properties cross validated with core sets ensure accuracy. To enhance permeability estimation, sophisticated machine learning (ML) methods employed, categorizing into five classes ranging from extremely good (0) very low (4). Two ML models, Naïve Bayes (NB) multilayer perceptron (MLP), applied predict permeability. The MLP model outperformed NB model, achieving 99% training accuracy 93% testing accuracy, compared 78 73%, respectively, model. resulting comprehensive revealed distribution across three stratigraphic layers: B100 zone exhibited suitable caprock, while D35-1 D35-2 zones demonstrated excellent indicating CO2 reservoirs. "X" field reservoir, located at depths exceeding 1300 m, meets depth requirements (1000–1500 m) storage. Our integrated approach, combining empirical ML-based calculations well logs, proved effective characterizing reservoir. lithological defined nonreservoir sections between clay silt lines, identifying important caprocks interbedded shale/clay intervals. Seismic profiling confirmed continuous caprock overlying D group zone, crucial preventing upward migration. This analysis supports Malay Basin viable storage, contributing ongoing efforts capture research.

Язык: Английский

Процитировано

0

Porosity prediction of tight reservoir rock using well logging data and machine learning DOI Creative Commons

Yawen He,

Hongjun Zhang,

Zhiyu Wu

и другие.

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

Опубликована: Апрель 16, 2025

Язык: Английский

Процитировано

0