Application of XGBoost model for in-situ water saturation determination in Canadian oil-sands by LF-NMR and density data

Markovic, Strahinja; Bryan, J.; Rezaee, Reza; Turakhanov, Aman; Kantzas, Apostolos; Koroteev, Dmitry

doi:10.1038/s41598-022-17886-6

Cited by 12 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… is the tree number, denotes the number of the leaf nodes, is the weight of each leaf, and are the regularization term. The regularization parameters determine the relative penalty of each term which is used to avoid the overfitting by controlling the complexity 53 – 55 . Large weights are penalized by the regularization term, which also encourages the model to have more streamlined and comprehensible structural elements 56 .…”

Section: Methodsmentioning

confidence: 99%

Detection of the chronic kidney disease using XGBoost classifier and explaining the influence of the attributes on the model using SHAP

Raihan

Khan

Kee

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Chronic kidney disease (CKD) is a condition distinguished by structural and functional changes to the kidney over time. Studies show that 10% of adults worldwide are affected by some kind of CKD, resulting in 1.2 million deaths. Recently, CKD has emerged as a leading cause of mortality worldwide, making it necessary to develop a Computer-Aided Diagnostic (CAD) system to diagnose CKD automatically. Machine Learning (ML) based CAD system can be used by a clinician to automatically diagnoses mass people. Since ML models are considered a black box, it is also necessary to expose influential causes behind a model's prediction of a particular output. So that, a doctor can make a more rational decision based on the model's output and analysis of the features influence on the model. In this paper, we have used the XGBoost as the ML classifier to predict whether a patient has CKD or not. Using the XGBoost classifier, we have obtained an accuracy, precision, recall, and F1 score of $$99.16{\%}, 100{\%}, 98.68{\%},$$ 99.16 % , 100 % , 98.68 % , and $$99.33{\%},$$ 99.33 % , respectively using all $$24$$ 24 features. Furthermore, we have used Biogeography Based Optimization (BBO) algorithm to find an effective subset of the features. The BBO algorithm selected almost half of the initial features. We have obtained an accuracy, precision, recall, and F1 score of $$98.33{\%}, 100{\%}, 97.36{\%},$$ 98.33 % , 100 % , 97.36 % , and $$98.67{\%},$$ 98.67 % , respectively using only 13 features selected by the BBO algorithm. Finally, we have explained the impact of the feature on the ML models using the SHapley Additive exPlanations (SHAP) analysis. Using SHAP analysis and BBO algorithm, we have found that hemoglobin and albumin mostly contribute to the detection of CKD.

show abstract

Section: Methodsmentioning

confidence: 99%

Detection of the chronic kidney disease using XGBoost classifier and explaining the influence of the attributes on the model using SHAP

Raihan

Khan

Kee

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…After the identification of such differences, the specific contact resistance was calculated by multiplying the jumps (in mΩ) by the specific contact area of the legs (in mm 2 ). The area of the TE legs was measured with the help of a common ruler (1 mm scale) and the error associated with the experimental set up was taken from the standard deviation of the fitted data “ROOT-MSE (SD)” tool [ 39 , 40 ], using the OriginPro software version 9.0. In each leg, the measurements were made along the two directions (first from left to right, L, and after from right to left, R), and often performed in more than one zone.…”

Section: Methodsmentioning

confidence: 99%

An Electrical Contacts Study for Tetrahedrite-Based Thermoelectric Generators

Coelho

Abreu²,

Carvalho

et al. 2022

Materials

View full text Add to dashboard Cite

High electrical and thermal contact resistances can ruin a thermoelectric device’s performance, and thus, the use of effective diffusion barriers and optimization of joining methods are crucial to implement them. In this work, the use of carbon as a Cu11Mn1Sb4S13 tetrahedrite diffusion barrier, and the effectiveness of different fixation techniques for the preparation of tetrahedrite/copper electrical contacts were investigated. Contacts were prepared using as jointing materials Ni and Ag conductive paints and resins, and a Zn-5wt% Al solder. Manual, cold- and hot-pressing fixation techniques were explored. The contact resistance was measured using a custom-made system based on the three points pulsed-current method. The legs interfaces (Cu/graphite/tetrahedrite) were investigated by optical and scanning electron microscopies, complemented with energy-dispersive X-ray spectroscopy, and X-ray diffraction. No interfacial phases were formed between the graphite and the tetrahedrite or Cu, pointing to graphite as a good diffusion barrier. Ag water-based paint was the best jointing material, but the use of hot pressing without jointing materials proves to be the most reliable technique, presenting the lowest contact resistance values. Computer simulations using the COMSOL software were performed to complement this study, indicating that high contact resistances strongly reduce the power output of thermoelectric devices.

show abstract

“…The study introduced a new empirical correlation derived from ANN biases and weights, validated with unseen data, providing a reliable method for Sw prediction without costly core analyses 42 . Markovic and colleagues (2022) conducted a thorough examination using the XGBoost technique to ascertain inter-pore SW within Canadian sandstone oil resources through the analysis of LF-NMR and density information 43 . Miah et al focus on the using of machine learning tools in estimating reservoir properties for hydrocarbon production through log-based reservoir characterization.…”

Section: Research Backgroundmentioning

confidence: 99%

Deep learning algorithm-enabled sediment characterization techniques to determination of water saturation for tight gas carbonate reservoirs in Bohai Bay Basin, China

Hu,

Meng,

Guo

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Understanding water saturation levels in tight gas carbonate reservoirs is vital for optimizing hydrocarbon production and mitigating challenges such as reduced permeability due to water saturation (Sw) and pore throat blockages, given its critical role in managing capillary pressure in water drive mechanisms reservoirs. Traditional sediment characterization methods such as core analysis, are often costly, invasive, and lack comprehensive spatial information. In recent years, several classical machine learning models have been developed to address these shortcomings. Traditional machine learning methods utilized in reservoir characterization encounter various challenges, including the ability to capture intricate relationships, potential overfitting, and handling extensive, multi-dimensional datasets. Moreover, these methods often face difficulties in dealing with temporal dependencies and subtle patterns within geological formations, particularly evident in heterogeneous carbonate reservoirs. Consequently, despite technological advancements, enhancing the reliability, interpretability, and applicability of predictive models remains imperative for effectively characterizing tight gas carbonate reservoirs. This study employs a novel data-driven strategy to prediction of water saturation in tight gas reservoir powered by three recurrent neural network type deep/shallow learning algorithms—Gated Recurrent Unit (GRU), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Support Vector Machine (SVM), K-nearest neighbor (KNN) and Decision tree (DT)—customized to accurately forecast sequential sedimentary structure data. These models, optimized using Adam's optimizer algorithm, demonstrated impressive performance in predicting water saturation levels using conventional petrophysical data. Particularly, the GRU model stood out, achieving remarkable accuracy (an R-squared value of 0.9973) with minimal errors (RMSE of 0.0198) compared to LSTM, RNN, SVM, KNN and, DT algorithms, thus showcasing its proficiency in processing extensive datasets and effectively identifying patterns. By achieving unprecedented accuracy levels, this study not only enhances the understanding of sediment properties and fluid saturation dynamics but also offers practical implications for reservoir management and hydrocarbon exploration in complex geological settings. These insights pave the way for more reliable and efficient decision-making processes, thereby advancing the forefront of reservoir engineering and petroleum geoscience.

show abstract

Application of XGBoost model for in-situ water saturation determination in Canadian oil-sands by LF-NMR and density data

Cited by 12 publications

References 23 publications

Detection of the chronic kidney disease using XGBoost classifier and explaining the influence of the attributes on the model using SHAP

Detection of the chronic kidney disease using XGBoost classifier and explaining the influence of the attributes on the model using SHAP

An Electrical Contacts Study for Tetrahedrite-Based Thermoelectric Generators

Deep learning algorithm-enabled sediment characterization techniques to determination of water saturation for tight gas carbonate reservoirs in Bohai Bay Basin, China

Contact Info

Product

Resources

About