Hydraulic flow unit and rock types of the Asmari Formation, an application of flow zone index and fuzzy C-means clustering methods

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Фев. 29, 2024

Rock types are the reservoir's most essential properties for special facies modeling in a defined range of porosity and permeability. This study used clustering techniques to identify rock 280 core samples from one wells drilled Asmari reservoir Mansouri field, SW Iran. Four hydraulic flow units (HFUs) were determined studied data utilizing histogram analysis, normal probability sum squared errors (SSE) statistical methods. Then, two zone index (FZI) fuzzy c-means (FCM) methods determine given well according results obtained HFU continuity acts in-depth. The FCM method, with number 3.12, compared FZI, 2.77, shows more depth. relationship between permeability improved considerably by techniques. improvement is achieved using FZI method study. Generally, all increased 0.55 0.81 first finally 0.94 fourth HFU. Similar an characterized samples. In comparison, correlation coefficients less than those general case HFUs. aims flowing fluid porous medium employing c-mean logic. Also, determining units, especially siliceous-clastic log Formation, third have highest quality Results can be nearby wellbores without cores.

Язык: Английский

---

Unsupervised machine learning-based multi-attributes analysis for enhancing gas channel detection and facies classification in the serpent field, offshore Nile Delta, Egypt

Geomechanics and Geophysics for Geo-Energy and Geo-Resources, Год журнала: 2024, Номер 10(1)

Опубликована: Ноя. 28, 2024

Abstract The prediction of highly heterogeneous reservoir parameters from seismic amplitude data is a major challenge. Seismic attribute analysis can enhance the tracking subtle stratigraphic features. It challenging to investigate these features, including channel systems, with conventional-amplitude data. Over past few years, use machine learning (ML) analyze multiple attributes has enhanced facies by mapping patterns in purpose this research was assess efficiency an unsupervised self-organizing map (SOM) approach supported multi-attribute that could improve gas detection and classification Serpent Field, offshore Nile Delta, Egypt. As well as evaluates importance several available characterization rather than analyzing individual volumes. In study, single (spectral decomposition attribute) highlighted spatial distribution using three distinct frequency magnitude values. Subsequently, we employ principal component (PCA) selection method, discovering combining such sweetness, envelope, spectral magnitude, voice input for SOM reflects effective method determine facies. clustering results distinguish between shale, shaly sand, wet gas-saturated sand identify gas–water contact on 2D topological (SOM), where each pattern indicates certain This achieved associating outputs lithofacies determined petrophysical logs. Reducing exploration development risk empowering geoscientist generate more precise interpretation are ultimate objectives analysis.

Язык: Английский

---

Enhancing formation bulk density prediction while drilling using mud logging data and interpretable boosting machine learning

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

Опубликована: Янв. 1, 2025

Язык: Английский

---

Improving permeability modeling using artificial neural networks in Lower Safa sand reservoir, JG field, Abu El-Gharadig basin, Western Desert, Egypt

Earth Science Informatics, Год журнала: 2025, Номер 18(2)

Опубликована: Янв. 31, 2025

Язык: Английский

---

Permeability, porosity, and water saturation relationships and distributions in complex reservoirs

Elsevier eBooks, Год журнала: 2025, Номер unknown, С. 151 - 185

Опубликована: Янв. 1, 2025

Язык: Английский

---

Extracting useful information from sparsely logged wellbores for improved rock typing of heterogeneous reservoir characterization using well-log attributes, feature influence and optimization

Petroleum Science, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

---

Machine learning with hyperparameter optimization applied in facies-supported permeability modeling in carbonate oil reservoirs

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

Опубликована: Апрель 15, 2025

Most carbonate reservoirs exhibit heterogeneous pore distribution, whereby the matrix displays low permeability, thus impeding flow of oil. On other hand, highly permeable fractures function as main conduits within such reservoirs. Permeability measurements are obtained from core and well test analysis, which too expensive not available for many wells. Therefore, accurate permeability prediction is a vital step in developing an efficient field development plan, it plays pivotal role distribution 3D petrophysical properties throughout reservoir. Machine learning (ML) algorithms now widely applied to predict using conventional logs build model uncored This review considers performance six ML (LightGBM, CATBoost, XGBoost, Adaboost, random forest gradient boosting) high-quality dataset. The dataset incorporates multiple well-log inputs (gamma ray, caliper, density, neutron porosity, shallow deep resistivity, total spontaneous potential, water saturation, depth, facies) addition direct porosity measurements. Data pre-processing techniques include missing data imputation, scale correction, normalization with three different transformations (log, Box-Cox, NST) outlier detection. To enhance performance, two search (random Bayesian optimization) compared their ability tune hyperparameters. There need identify suitable parameter space, especially when target variable range changing. was evaluated four evaluation metrics (RMSE, MAE, R2, Adjusted R2). Results showed that XGBoost algorithm configuration (RS algorithm, Box Cox method, Z-score detection, without old space) delivered best RMSE values 6.9 md 9.78 training testing, respectively.

Язык: Английский

---

Leveraging Boosting Machine Learning for Drilling Rate of Penetration (ROP) Prediction Based on Drilling and Petrophysical Parameters

Artificial Intelligence in Geosciences, Год журнала: 2025, Номер unknown, С. 100121 - 100121

Опубликована: Май 1, 2025

Язык: Английский

---

Improving permeability prediction via Machine Learning in a heterogeneous carbonate reservoir: application to Middle Miocene Nullipore, Ras Fanar field, Gulf of Suez, Egypt

Environmental Earth Sciences, Год журнала: 2024, Номер 83(8)

Опубликована: Апрель 1, 2024

Abstract Predicting and interpolating the permeability between wells to obtain 3D distribution is a challenging mission in reservoir simulation. The high degree of heterogeneity diagenesis Nullipore carbonate provide significant obstacle accurate prediction. Moreover, intricate relationships core well logging data exist reservoir. This study presents novel approach based on Machine Learning (ML) overcome such difficulties build robust predictive model. main objective this develop an ML-based prediction predict logs populate predicted methodology involves grouping cored intervals into flow units (FUs), each which has distinct petrophysical characteristics. probability density function used investigate FUs select high-weighted input features for reliable model Five ML algorithms, including Linear Regression (LR), Polynomial (PR), Support Vector (SVR), Decision Trees (DeT), Random Forests (RF), have been implemented integrate with influential permeability. dataset randomly split training testing sets evaluate performance developed models. models’ hyperparameters were tuned improve model’s performance. To logs, two key containing whole are train most model, other test Results indicate that RF outperforms all models offers results, where adjusted coefficient determination ( R 2 adj ) 0.87 set 0.82 set, mean absolute error squared (MSE) 0.32 0.19, respectively, both sets. It was observed exhibits when it trained FUs. aids detecting patterns along profile capturing wide Ultimately, populated via Gaussian Function Simulation geostatistical method outcomes will aid users make informed choices appropriate algorithms use characterization more predictions better decision-making limited available data.

Язык: Английский

---

An interpretable ensemble machine-learning workflow for permeability predictions in tight sandstone reservoirs using logging data

Geophysics, Год журнала: 2024, Номер 89(5), С. MR265 - MR280

Опубликована: Май 30, 2024

Tight sandstone reservoirs exhibit strong vertical heterogeneity and complex pore structures, challenging conventional permeability evaluation methods based on well-logging data. Although rising machine-learning (ML) techniques have demonstrated excellent accuracy for industrial applications, the physics rationality within such a powerful “black box” remain less clear. Hence, reliable prediction would benefit from an interpretable ML-based workflow that could reveal controlling factors. To compare models examine underlying features, 16 different ML submodels are tested after data preprocessing, feature selection, hyperparameter optimization. By comparing fitting tuning time, light gradient boosting machine optimized by whale optimization algorithm, referred to as LGB-WOA, is determined be optimal model with best relatively short time. A field application demonstrates even in highly heterogeneous reservoir sections, LGB-WOA outperformed petrophysical being most consistent directly measured core samples ([Formula: see text]). The Shapley additive explanation values then used interpret predictions of our model. As expected, porosity curve exhibits highest importance among all input significantly contributing predictions. Conversely, wellbore diameter compensated neutron log contribute least need not subsequent improvements. These experiments provide method accurately assessing broader understanding characterization, paving way establishing more models.

Язык: Английский

---