Machine learning analyses of low salinity effect due to surface-charge alteration in carbonate porous media

UCS Prediction of Carbonate Rocks using Deep Feedforward Neural Networks with Aquila Algorithm Model

For safe and cost effective design of geotechnical and rock engineering schemes, accurate Uniaxial Compressive Strength (UCS) prediction is crucial, especially in carbonate rock formations with considerable heterogeneity. For enhancing the precision and resilience of UCS prediction, this study proposes a hybrid intelligent modelling framework that combines Aquila Optimization (AO) with a Deep Feedforward Neural Network (DFNN). Comprehensive data collected from laboratory-tested carbonate rock samples are preprocessed, which includes handling missing values and data cleaning. To comprehend the behavior and correlations of input variables, exploratory data visualization and feature distribution analysis are carried out. The regression model is a deep feedforward neural network and important hyperparameters like number of hidden layers, neurons, learning rate and activation functions are optimally tuned using Aquila Optimisation Algorithm (AOA). Multiple regression metrics are used to quantitatively assess model performance from python software. The findings show that proposed framework attains minimum RMSE and MSE value of 0.0425,0.0018 is useful for rock mechanics and mining applications and gives a dependable, data-driven tool for UCS prediction in carbonate rocks.

Predictive modeling of groundwater quality for irrigation using integrated indices and machine learning: case study in Tunisia

Research on a machine learning-based prediction method for methane adsorption capacity in shale

Accurate prediction of shale methane adsorption capacity is crucial for reservoir evaluation. This study integrates 486 experimental datasets to develop a multivariate machine learning prediction model. Six key geological parameters, including depth, total organic carbon (TOC), moisture, porosity, vitrinite reflectance ( Ro ), and clay minerals, were selected as features. Correlation analysis methods were used to reveal the nonlinear relationships between various geological parameters and methane adsorption capacity, as well as the intrinsic coupling structures among these parameters. The predictive performance of three machine learning models, namely, Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), was systematically compared. The results show that TOC and Ro are the significant factors controlling methane adsorption. The XGBoost model achieved superior performance on both training and test sets, demonstrating higher prediction accuracy and stronger generalization capability compared to SVM and RF. This study provides novel insights and methodologies for assessing shale gas adsorption potential and optimizing development strategies.

Prediction of Porosity and Permeability Using Well Log and Core Data: A Data-Driven Approach

Accurate prediction of porosity and permeability is very important for reservoir characterization and hydrocarbon extraction. Traditional workflow in the form of empirical correlations is usually difficult, time-consuming, spatially limiting, and entirely dependent on formation geology. The current study examines a different approach, which uses machine learning (ML) regression models based on well-log and core data. Three regression architectures including Random Forest, CatBoost, and K-Nearest Neighbors (KNN) were trained and validated based on a dataset consisting of 340 samples of shaly sand gas reservoirs. The gamma ray (GR), resistivity (RLLD), spontaneous potential (SP), bulk density (RHOB), neutron porosity (NPHI), and depth were used as the input variables with the core-derived porosity (CPHI) and permeability (CKHG) being used as the targets. The quantitative measures of performance of the models included R2, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). The findings indicated that KNN regression was better than its counterparts as it achieved R2 = 0.8933 and R2 = 0.9340 in terms of porosity and permeability prediction, respectively, and more acceptable metrics of errors showed. Comparatively, the traditional empirical methods showed a significantly lower accuracy rate. The findings highlight that machine learning has the potential to provide precise, scalable, and low-cost predictions of the reservoir properties which could lead to better choices for exploration and production activities.

Comparative Analysis of Five Machine Learning Regression Models for Salinity Prediction in the California Current System

Salinity significantly influences ocean water movement, climate change, marine ecosystems and heat transfer, impacting processes such as ocean circulation patterns, density-driven currents, sea ice production, nutrient distribution, and the global climate system. Various physics-based equations and oceanographic theories are utilized to measure saltwater salinity in specific locations. This approach fails to account for data fluctuation, harsh environmental conditions, and the detection of oceanographic data patterns. This type of salinity measurement is time-consuming and expensive. This study evaluates five superior regression models: multivariable linear regression (MVR), K nearest neighbor (KNN), random forest (RF), support vector regression (SVR), and extreme gradient boosting (XGBoost). After performing five regression models, the XGBoost model demonstrated superior performance compared to the other four models with the highest coefficient of determination (R²). The score recorded was 0.9981, root mean square error (RMSE): 0.0232, mean square error (MSE): 0.0005, and mean absolute error (MAE): 0.0152. Cross- validation was also conducted for the best-performing model for generalization of model output. This study provides a robust and cost-efficient tool and proves the efficiency of the ML model for measuring seawater salinity.

Sustainable mineral processing risk analysis based initial settling rates prediction using enhanced machine learning models

The primary contamination concern of mineral processing tailings (MPT) is the leaching of hazardous substances into the environment. The literature indicates that MPT requires effective flocculation and polymer-assisted dewatering to ensure its disposal does not cause environmental damage. In this research, a theoretical modeling framework was adopted, based on the development of a hybrid machine learning (ML) model for predicting flocculation-dewatering efficiency, aiming to reduce the cost of laboratory tests. The proposed ML model is based on an efficient Gaussian process regression (GPR) model that uses a feature impact strategy (FIS) and a kernel matrix dynamically updated using error noise, named enhanced GPR (EGPR). Additionally, the kernel ridge model and SHAP (SHapley Additive exPlanations) method are coupled for feature selection (FS), identifying the most important features among 17 input variables (features). The target variable in this research is the initial settling rate (ISR), which is predicted to utilize the EGPR model. The statistical analysis revealed that the EGPR model outperforms the Deep random vector functional link (DRVFL), least square support vector machine (LSSVM), cascade feedforward neural network (CFNN), and ridge regression with superior error metrics (R = 0.951, RMSE = 0.196, MAPE = 62.22). It also demonstrated the least uncertainty (UI = 17.65), which indicates its reliability and accuracy. Risk analysis (RA) indicates that the EGPR model yields the lowest total risk score (TRS) (5.6) and is classified as a “Very Low” risk method. Furthermore, SHAP analysis exhibits that the solids content (SC) and flocculant dose (FD) positively influenced the prediction of ISR. Consequently, this study presents a foundational methodology for predicting ISR, which could be introduced as an essential tool for flocculate-settling studies that contribute to optimal chemical dosage, real-time contamination monitoring, and the operation of water recovery storage capacity.