{"title":"A novel approach to classify lithology of reservoir formations using GrowNet and Deep-Insight with physic-based feature augmentation","authors":"Seyed Hamid Reza Mousavi, Seyed Mojtaba Hosseini-Nasab","doi":"10.1002/ese3.1895","DOIUrl":null,"url":null,"abstract":"<p>Manual interpretation of geophysical logging data can be a tedious and time-consuming task in the case of the nonlinear behavior of well-logging signals. This study aims to enhance lithology classification of reservoir formations through advanced machine learning (ML) and deep learning (DL) techniques, introducing and comparing three novel algorithms, GrowNet, Deep-Insight, and blender, against traditional models like random forest (RF) and support vector machine (SVM). Data from the South and North Viking Graben regions, encompassing 12 lithological facies, was preprocessed through cleaning, normalization, transformation, and imputation of missing values using regression models. The data set was enhanced with physic-based features and balanced using SMOTE and NearMiss algorithms. Deep-Insight converted tabular data into images for a convolutional neural network (CNN), significantly improving classification accuracy compared to conventional models such as decision trees (DTs). GrowNet and blender models leveraged hybrid approaches for enhanced performance. These hybrid approaches successfully addressed data imbalance and enhanced model learning, outperforming classic methods. The GrowNet and blender models for lithology classification successfully increased the penalty score and accuracy compared to the FORCE 2020 competition. Additionally, introducing the class prediction error plot visualizes multiclass classification performance more effectively than using a confusion matrix. These novel models in multiclass classification contribute to the petroleum industry by providing more accurate and reliable lithology classification, thereby improving reservoir characterization and exploration efficiency.</p>","PeriodicalId":11673,"journal":{"name":"Energy Science & Engineering","volume":null,"pages":null},"PeriodicalIF":3.5000,"publicationDate":"2024-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ese3.1895","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy Science & Engineering","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/ese3.1895","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0
Abstract
Manual interpretation of geophysical logging data can be a tedious and time-consuming task in the case of the nonlinear behavior of well-logging signals. This study aims to enhance lithology classification of reservoir formations through advanced machine learning (ML) and deep learning (DL) techniques, introducing and comparing three novel algorithms, GrowNet, Deep-Insight, and blender, against traditional models like random forest (RF) and support vector machine (SVM). Data from the South and North Viking Graben regions, encompassing 12 lithological facies, was preprocessed through cleaning, normalization, transformation, and imputation of missing values using regression models. The data set was enhanced with physic-based features and balanced using SMOTE and NearMiss algorithms. Deep-Insight converted tabular data into images for a convolutional neural network (CNN), significantly improving classification accuracy compared to conventional models such as decision trees (DTs). GrowNet and blender models leveraged hybrid approaches for enhanced performance. These hybrid approaches successfully addressed data imbalance and enhanced model learning, outperforming classic methods. The GrowNet and blender models for lithology classification successfully increased the penalty score and accuracy compared to the FORCE 2020 competition. Additionally, introducing the class prediction error plot visualizes multiclass classification performance more effectively than using a confusion matrix. These novel models in multiclass classification contribute to the petroleum industry by providing more accurate and reliable lithology classification, thereby improving reservoir characterization and exploration efficiency.
期刊介绍:
Energy Science & Engineering is a peer reviewed, open access journal dedicated to fundamental and applied research on energy and supply and use. Published as a co-operative venture of Wiley and SCI (Society of Chemical Industry), the journal offers authors a fast route to publication and the ability to share their research with the widest possible audience of scientists, professionals and other interested people across the globe. Securing an affordable and low carbon energy supply is a critical challenge of the 21st century and the solutions will require collaboration between scientists and engineers worldwide. This new journal aims to facilitate collaboration and spark innovation in energy research and development. Due to the importance of this topic to society and economic development the journal will give priority to quality research papers that are accessible to a broad readership and discuss sustainable, state-of-the art approaches to shaping the future of energy. This multidisciplinary journal will appeal to all researchers and professionals working in any area of energy in academia, industry or government, including scientists, engineers, consultants, policy-makers, government officials, economists and corporate organisations.