Performance evaluation of boosting machine learning algorithms for lithofacies classification in heterogeneous carbonate reservoirs

Al‐Mudhafar, Watheq J.; Abbas, Mohammed A.; Wood, David A.

doi:10.1016/j.marpetgeo.2022.105886

Cited by 35 publications

(18 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Al-Mudhafar et al used the genetic algorithm to predict the changing pattern of the bottom-hole pressure during the production of horizontal wells. The error level was reduced to 10% compared with the measured data. − Although more machine learning methods are applied to the dynamic production capacity analysis of tight reservoirs, the different methods cannot accurately predict the recovery due to the differences in evaluation indexes, data characteristics, and processing methods. There is an urgent need to propose a new model for capacity prediction that integrates multiple influencing factors to predict recovery in tight reservoirs. − …”

Section: Introductionmentioning

confidence: 99%

New Model for Predicting Production Capacity of Horizontal Well Volume Fracturing in Tight Reservoirs

Lei,

Daihong

2024

ACS Omega

View full text Add to dashboard Cite

Production prediction is the most important and comprehensive index to measure the effect of oilfield development, and it is also one of the most fundamental problems in oilfield dynamic analysis. However, the recovery prediction is often affected by many factors. Usually, the recovery is predicted by core experiments, numerical simulations, and mathematical models. The main problem is accurately predicting reservoir recovery based on existing data. This paper proposes a comprehensive prediction model for the problem of recovery. First, the correlation coefficients between 14 factors and recovery were calculated based on Pearson, Spearman, gray correlation, variance selection, univariate selection method, and tree model. Second, the weights of the factors were determined using entropy weighting, CRITIC, and hierarchical analysis to clarify the degree of contribution of different factors to the recovery. Finally, a comprehensive evaluation model was established based on the results of the weighting analysis. The results indicate that the correlation coefficient and weight of porosity, permeability, oil saturation, well spacing, cluster spacing, total fluid volume, and horizontal section length are the most relevant to the recovery. The error between the comprehensive evaluation model and the actual results is less than 3%. Therefore, the method can predict the production capacity of the tight reservoirs. The research results of this paper are of guiding significance for improving the recovery of tight reservoirs.

show abstract

Section: Introductionmentioning

confidence: 99%

New Model for Predicting Production Capacity of Horizontal Well Volume Fracturing in Tight Reservoirs

Lei,

Daihong

2024

ACS Omega

View full text Add to dashboard Cite

show abstract

“…Supervised learning methods are more concerned with the correlations between geological properties and logging responses. These include Bayesian inversion (Qin et al, 2018;Feng, 2021), decision trees (Ren et al, 2022), support vector machines (SU et al, 2020), neural networks (Gu et al, 2019), gradient boosting algorithms (Gu et al, 2021;Al-Mudhafar et al, 2022;Zheng et al, 2022), random forests (Antariksa et al, 2022), and emerging deep learning methods (Song et al, 2020;Liu and Liu, 2022). Among these methods, Bayesian inversion can apply different prior frameworks and likelihood models to avoid inappropriate transitions among different lithofacies in geology and petrophysics (Hammer et al, 2012).…”

Section: Introductionmentioning

confidence: 99%

Lithofacies logging identification for strongly heterogeneous deep-buried reservoirs based on improved Bayesian inversion: The Lower Jurassic sandstone, Central Junggar Basin, China

Zheng

Zhang²,

Cheng³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

The strong heterogeneity characteristics of deep-buried clastic low-permeability reservoirs may lead to great risks in hydrocarbon exploration and development, which makes the accurate identification of reservoir lithofacies crucial for improving the obtained exploration results. Due to the very limited core data acquired from deep drilling, lithofacies logging identification has become the most important method for comprehensively obtaining the rock information of deep-buried reservoirs and is a fundamental task for carrying out reservoir characterization and geological modeling. In this study, a machine learning method is introduced to lithofacies logging identification, to explore an accurate lithofacies identification method for deep fluvial-delta sandstone reservoirs with frequent lithofacies changes. Here Sangonghe Formation in the Central Junggar Basin of China is taken as an example. The K-means-based synthetic minority oversampling technique (K-means SMOTE) is employed to solve the problem regarding the imbalanced lithofacies data categories used to calibrate logging data, and a probabilistic calibration method is introduced to correct the likelihood function. To address the situation in which traditional machine learning methods ignore the geological deposition process, we introduce a depositional prior for controlling the vertical spreading process based on a Markov chain and propose an improved Bayesian inversion process for training on the log data to identify lithofacies. The results of a series of experiments show that, compared with the traditional machine learning method, the new method improves the recognition accuracy by 20%, and the predicted petrographic vertical distribution results are consistent with geological constraints. In addition, SMOTE and probabilistic calibration can effectively handle data imbalance problems so that different categories can be adequately learned. Also the introduction of geological prior has a positive impact on the overall distribution, which significantly improves the accuracy and recall rate of the method. According to this comprehensive analysis, the proposed method greatly enhanced the identification of the lithofacies distributions in the Sangonghe Formation. Therefore, this method can provide a tool for logging lithofacies interpretation of deep and strongly heterogeneous clastic reservoirs in fluvial-delta and other depositional environments.

show abstract

“…However, decision tree algorithms have a large memory footprint and are less efficient at processing many variables, which can lead to overfitting. 18 Therefore, a new algorithm, LightGBM, was developed based on XGBoost. It has a small memory footprint, efficient training speed, accurate prediction ability, can handle large-scale data and high-dimensional features, and provides flexible parameter adjustment and parallelization functions, which can effectively improve model performance and efficiency.…”

Section: Introductionmentioning

confidence: 99%

“…Sun et al 17 experimentally demonstrated that the XGBoost decision tree model performed well in lithology prediction. However, decision tree algorithms have a large memory footprint and are less efficient at processing many variables, which can lead to overfitting 18 . Therefore, a new algorithm, LightGBM, was developed based on XGBoost.…”

Section: Introductionmentioning

confidence: 99%

Lithofacies identification of shale formation based on mineral content regression using LightGBM algorithm: A case study in the Luzhou block, South Sichuan Basin, China

Liu,

Zhu,

Zhai

et al. 2023

Energy Science & Engineering

View full text Add to dashboard Cite

Lithofacies form the basis for evaluating shale gas fields and play an important role in gas reservoir enrichment. The accurate identification of shale lithofacies is key for exploration and development. Based on well‐logged data, the accuracy of mineral content prediction using machine‐learning regression models is not ideal. Therefore, feature derivation was introduced to enhance the correlation between minerals and lithofacies and improve the data expression ability. Four machine‐learning models for mineral regression were established based on feature‐derived data sets: LightGBM, XGBoost, artificial neural network, and support vector machine. By calculating the evaluation metrics of each model, we found that LightGBM had the best prediction performance. To compare and confirm the accuracy of the model in identifying lithofacies, this study established a new method, MT‐LightGBM, which combines the LightGBM mineral content regression model with mineral ternary diagrams to identify lithofacies. By using the MT‐LightGBM model and LightGBM classification models to identify the target lithofacies, it was found that the accuracy of lithofacies identification of MT‐LightGBM reached 94%. This accuracy is high and is of great significance for understanding and evaluating underground shale reservoirs.

show abstract

Performance evaluation of boosting machine learning algorithms for lithofacies classification in heterogeneous carbonate reservoirs

Cited by 35 publications

References 39 publications

New Model for Predicting Production Capacity of Horizontal Well Volume Fracturing in Tight Reservoirs

New Model for Predicting Production Capacity of Horizontal Well Volume Fracturing in Tight Reservoirs

Lithofacies logging identification for strongly heterogeneous deep-buried reservoirs based on improved Bayesian inversion: The Lower Jurassic sandstone, Central Junggar Basin, China

Lithofacies identification of shale formation based on mineral content regression using LightGBM algorithm: A case study in the Luzhou block, South Sichuan Basin, China

Contact Info

Product

Resources

About