Intelligent Logging Lithological Interpretation With Convolution Neural Networks

Zhu, Lei; Li, Hongqi; Yang, Zhongguo; Li, Chengyang; Ao, Yile

doi:10.30632/pjv59n6-2018a5

Cited by 19 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work mainly focuses on multi-class classification, which might result in lower prediction accuracy. Zhu et al (2018) utilized the waveletdecomposition approach to convert the lithology identification task into a supervised image recognition task. A convolution neural network (CNN) was then used to train the model to classify lithology classes.…”

Section: Related Workmentioning

confidence: 99%

A Coarse-to-Fine Approach for Intelligent Logging Lithology Identification with Extremely Randomized Trees

Xie

Zhu

et al. 2020

Math Geosci

View full text Add to dashboard Cite

Lithology identification is vital for reservoir exploration and petroleum engineering. Recently, there has been growing interest in using an intelligent logging approach for lithology classification. Machine learning has emerged as a powerful tool in inferring lithology types with the logging curves. However, well logs are susceptible to logging parameter manual entry, borehole conditions and tool calibrations. Most studies in the field of lithology classification with machine learning approaches have focused only on improving the prediction accuracy of classifiers. Also, a model trained in one location is not reusable in a new location due to different data distributions. In this paper, a unified framework is provided for training a multi-class lithology classification model for a data set with outlier data. In this paper, a coarse-to-fine framework that combines outlier detection, multi-class classification with an extremely randomized tree-based classifier is proposed to solve these issues. An unsupervised learning approach is used to detect the outliers in the data set. Then a coarse-to-fine inference procedure is used to infer the lithology class with an extremely randomized tree classifier. Two real-world data sets of well-logging are used to demonstrate the effectiveness of the proposed framework. Comparisons are conducted with some baseline machine learning classifiers, namely random forest, gradient tree boosting, and xgboosting. Results show that the proposed framework has higher prediction accuracy in sandstones compared with other approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

A Coarse-to-Fine Approach for Intelligent Logging Lithology Identification with Extremely Randomized Trees

Xie

Zhu

et al. 2020

Math Geosci

View full text Add to dashboard Cite

show abstract

“…Compared to petrophysical model-based methods, e.g., Timur-Coats equation and Windland's equation (Leverett, 1941;Timur, 1968), machine-learning approaches obtain better results by fitting more complex relationships between well logs and formation properties. Common machinelearning methods used for well-log interpretation include random forest, feed-forward neural network, and convolutional neural network (CNN) (Collobert and Weston, 2008;Bhattacharya and Mishra., 2018;Zhu et al, 2018;Zhong et al, 2019). However, in almost all previous research, the training data set is formed by lumping data from all available wells and is not adapted to the test set, which implicitly assumes that the measurement error of well logs is stationary, i.e., the statistics of the error do not vary with well locations or depths, and the welllog interpretation is unique, i.e., formation properties can be uniquely determined from a set of well logs.…”

Section: Introductionmentioning

confidence: 99%

Reducing the Uncertainty of Multi-Well Petrophysical Interpretation from Well Logs via Machine-Learning and Statistical Models

Pan¹,

Torres‐Verdín²,

Duncan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Well-log interpretation provides in situ estimates of formation properties such as porosity, hydrocarbon pore volume, and permeability. Reservoir models based on well-log-derived formation properties deliver reserve-volume estimates, production forecasts, and help with decision making in reservoir development. However, due to measurement errors, variability of well logs due to multiple measurement vendors, different borehole tools, and non-uniform drilling/borehole conditions, conventional well-log interpretation methods may not yield accurate estimates of formation properties, especially in the context of multi-well interpretation. To improve the robustness of multi-well petrophysical interpretation, well-log normalization techniques such as two-point scaling and mean-variance normalization are commonly used to impose stationarity constraints for well logs requiring correction. However, these techniques are mostly based on the marginal distribution of well logs and require expert knowledge to be effectively implemented. To reduce the uncertainties and time associated with multi-well petrophysical interpretation, we develop the discriminative adversarial (DA) model and the linear constraint model for well-log normalization and interpretation. We also develop a new divergence-based type well identification method for improved test-well and training-well adaptation.The DA neural network model developed for well-log normalization and interpretation can perform both linear and nonlinear well-log normalization by considering the joint distribution of all types of well logs and formation properties. To train the DA model, classical machine-learning models or classical petrophysical models are first trained to minimize the prediction error of formation properties in the training data set; then the adversarial model is trained to normalize well logs in the test set, such that the joint distribution of normalized well logs and formation property estimates of the test data set reproduce those of the training data set. The linear constraint model uses an ensemble of predictions from linear models to constrain both well-log normalization and interpretation. To identify wells with stationary formation properties as well as well logs, the divergence-based type well identification method is developed to choose type wells for wells requiring correction based on well-log statistical similarity instead of closeness of wells.We apply the developed methods to improve the accuracy of well-log normalization and the estimation of permeability in a carbonate reservoir. Six types of well logs and over 9000 feet of core measurements from 30 wells drilled between 1980s and 2010s in the Seminole San Andres Unit are available to validate the new multi-well interpretation workflow. Our interpretation models is flexible to integrate any types of classical machine-learning methods and petrophysical assumptions for robust petrophysical estimations. In comparison to classical machine-learning models with no normalization, with two-point scaling normalization and with linear constraints, the DA method yields better performance, e.g., the mean-squared error of permeability estimation decreases by approximately 20-50%. Our interpretation workflow can be applied to other stationary signal and image processing problems to mitigate errors introduced by biased measurements, and to better adapt models calibrated with data from one field to other neighboring fields.

show abstract

“…Therefore, Saporetti et al [ 18 ] adopted an evolutionary parameter tuning strategy, and combined gradient boosting (GB) with differential evolution (DE) to achieve the optimization of super parameters, thereby making the lithology identification more stable. In the study of [ 19 ], the wavelet decomposition was used to construct multi-channel images of logging data, and then the lithology identification problem based on the logging curve was skillfully transformed into an image segmentation problem. Finally, the feasibility of this method was verified by the application in the Daqing oilfield.…”

Section: Introductionmentioning

confidence: 99%

Well Logging Based Lithology Identification Model Establishment Under Data Drift: A Transfer Learning Method

Liu

Cao

et al. 2020

Sensors

View full text Add to dashboard Cite

Recent years have witnessed the development of the applications of machine learning technologies to well logging-based lithology identification. Most of the existing work assumes that the well loggings gathered from different wells share the same probability distribution; however, the variations in sedimentary environment and well-logging technique might cause the data drift problem; i.e., data of different wells have different probability distributions. Therefore, the model trained on old wells does not perform well in predicting the lithologies in newly-coming wells, which motivates us to propose a transfer learning method named the data drift joint adaptation extreme learning machine (DDJA-ELM) to increase the accuracy of the old model applying to new wells. In such a method, three key points, i.e., the project mean maximum mean discrepancy, joint distribution domain adaptation, and manifold regularization, are incorporated into extreme learning machine. As found experimentally in multiple wells in Jiyang Depression, Bohai Bay Basin, DDJA-ELM could significantly increase the accuracy of an old model when identifying the lithologies in new wells.

show abstract

Intelligent Logging Lithological Interpretation With Convolution Neural Networks

Cited by 19 publications

References 0 publications

A Coarse-to-Fine Approach for Intelligent Logging Lithology Identification with Extremely Randomized Trees

A Coarse-to-Fine Approach for Intelligent Logging Lithology Identification with Extremely Randomized Trees

Reducing the Uncertainty of Multi-Well Petrophysical Interpretation from Well Logs via Machine-Learning and Statistical Models

Well Logging Based Lithology Identification Model Establishment Under Data Drift: A Transfer Learning Method

Contact Info

Product

Resources

About