Quantitative uncertainty evaluation of seismic facies classification: A case study from northeast China

Geophysics ◽  
2016 ◽  
Vol 81 (3) ◽  
pp. B87-B99 ◽  
Author(s):  
Cheng Yuan ◽  
Jingye Li ◽  
Xiaohong Chen ◽  
Ying Rao
Keyword(s):  
Seismic Data ◽  
Northeast China ◽  
Reservoir Characterization ◽  
Early Stage ◽  
Uncertainty Propagation ◽  
Seismic Facies ◽  
Uncertainty Evaluation ◽  
Computational Time ◽  
Facies Classification ◽  
Well Data

Reservoir characterization in the early stage of oilfield exploration generally has enormous uncertainty because few geophysical and well data are typically available. The uncertainty when classifying the facies with seismic data propagates throughout the processes of seismic facies classification, causing errors in the final evaluation of geologic features in an area. To quantitatively evaluate the uncertainty in seismic facies classification, we have analyzed prestack seismic data and well observations in a tight reservoir from northeast China and calculated the uncertainties throughout the process. To achieve this, the facies probabilities conditioned on different properties in each step of seismic facies classification were first derived using a probabilistic multistep inversion. Second, the associated uncertainty and maximum a posterior (MAP) of facies probabilities were evaluated by means of entropy and reconstruction rate, which assessed the degree of similarity between MAP and facies sequence within the range [0, 1]. This enabled us to investigate the influence of the uncertainty propagation on seismic facies classification. The uncertainty of the inversion results for the target reservoir was finally characterized by the calculated entropy and its indicator transform. Additionally, parameter spaces of well-log and upscaled elastic properties were restricted according to the data distribution characteristics in the crossplot. Parameter vectors that were outside the restricted scopes were excluded, reducing the computational time and uncertainty. We determined that quantitative uncertainty evaluation by entropy with a probabilistic multistep approach enabled us to explore much more details of the uncertainty propagation in the processes of seismic reservoir characterization. It should be the method of choice for risk of management and decision making in reservoir assessment.

Applications of Artificial Neural Networks for Seismic Facies Classification

2021 ◽  
Vol 73 (02) ◽  
pp. 68-69
Author(s):  
Chris Carpenter
Keyword(s):  
Neural Network ◽  
Seismic Data ◽  
Reservoir Characterization ◽  
Probabilistic Neural Network ◽  
Oil Field ◽  
Seismic Facies ◽  
Data Set ◽  
Full Field ◽  
Facies Classification

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 200577, “Applications of Artificial Neural Networks for Seismic Facies Classification: A Case Study From the Mid-Cretaceous Reservoir in a Supergiant Oil Field,” by Ali Al-Ali, Karl Stephen, SPE, and Asghar Shams, Heriot-Watt University, prepared for the 2020 SPE Europec featured at the 82nd EAGE Conference and Exhibition, originally scheduled to be held in Amsterdam, 1-3 December. The paper has not been peer reviewed. Facies classification using data from sources such as wells and outcrops cannot capture all reservoir characterization in the interwell region. Therefore, as an alternative approach, seismic facies classification schemes are applied to reduce the uncertainties in the reservoir model. In this study, a machine-learning neural network was introduced to predict the lithology required for building a full-field Earth model for carbonate reservoirs in southern Iraq. The work and the methodology provide a significant improvement in facies classification and reveal the capability of a probabilistic neural network technique. Introduction The use of machine learning in seismic facies classification has increased gradually during the past decade in the interpretation of 3D and 4D seismic volumes and reservoir characterization work flows. The complete paper provides a literature review regarding this topic. Previously, seismic reservoir characterization has revealed the heterogeneity of the Mishrif reservoir and its distribution in terms of the pore system and the structural model. However, the main objective of this work is to classify and predict the heterogeneous facies of the carbonate Mishrif reservoir in a giant oil field using a multilayer feed-forward network (MLFN) and a probabilistic neural network (PNN) in nonlinear facies classification techniques. A related objective was to find any domain-specific causal relationships among input and output variables. These two methods have been applied to classify and predict the presence of different facies in Mishrif reservoir rock types. Case Study Reservoir and Data Set Description. The West Qurna field is a giant, multibillion-barrel oil field in the southern Mesopotamian Basin with multiple carbonate and clastic reservoirs. The overall structure of the field is a north/south trending anticline steep on the western flank and gentle on the eastern flank. Many producing reservoirs developed in this oil field; however, the Mid- Cretaceous Mishrif reservoir is the main producing reservoir. The reservoir consists of thick carbonate strata (roughly 250 m) deposited on a shallow water platform adjacent to more-distal, deeper-water nonreservoir carbonate facies developing into three stratigraphic sequence units in the second order. Mishrif facies are characterized by a porosity greater than 20% and large permeability contrast from grainstones to microporosity (10-1000 md). The first full-field 3D seismic data set was achieved over 500 km2 during 2012 and 2013 in order to plan the development of all field reservoirs. A de-tailed description of the reservoir has been determined from well logs and core and seismic data. This study is mainly based on facies log (22 wells) and high-resolution 3D seismic volume to generate seismic attributes as the input data for the training of the neural network model. The model is used to evaluate lithofacies in wells without core data but with appropriate facies logs. Also, testing was carried out in parallel with the core data to verify the results of facies classification.

Seismic facies classification using supervised convolutional neural networks and semisupervised generative adversarial networks

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. O47-O58 ◽  
Author(s):  
Mingliang Liu ◽  
Michael Jervis ◽  
Weichang Li ◽  
Philippe Nivlet
Keyword(s):  
Neural Networks ◽  
Seismic Data ◽  
Hydrocarbon Exploration ◽  
Training Data ◽  
Seismic Facies ◽  
Generative Adversarial Networks ◽  
Complex Data ◽  
Facies Classification ◽  
Good Ability ◽  
Well Data

Mapping of seismic and lithologic facies from 3D reflection seismic data plays a key role in depositional environment analysis and reservoir characterization during hydrocarbon exploration and development. Although a variety of machine-learning methods have been developed to speed up interpretation and improve prediction accuracy, there still exist significant challenges in 3D multiclass seismic facies classification in practice. Some of these limitations include complex data representation, limited training data with labels, imbalanced facies class distribution, and lack of rigorous performance evaluation metrics. To overcome these challenges, we have developed a supervised convolutional neural network (CNN) and a semisupervised generative adversarial network (GAN) for 3D seismic facies classification in situations with sufficient and limited well data, respectively. The proposed models can predict 3D facies distribution based on actual well log data and core analysis, or other prior geologic knowledge. Therefore, they provide a more consistent and meaningful implication to seismic interpretation than commonly used unsupervised approaches. The two deep neural networks have been tested successfully on a realistic synthetic case based on an existing reservoir and a real case study of the F3 seismic data from the Dutch sector of the North Sea. The prediction results show that, with relatively abundant well data, the supervised CNN-based learning method has a good ability in feature learning from seismic data and accurately recovering the 3D facies model, whereas the semisupervised GAN is effective in avoiding overfitting in the case of extremely limited well data. The latter seems, therefore, particularly adapted to exploration or early field development stages in which labeled data from wells are still very scarce.

Multiattribute framework analysis for the identification of carbonate mounds in the Brazilian presalt zone

2019 ◽  
Vol 7 (2) ◽  
pp. T467-T476 ◽  
Author(s):  
Carlos Jesus ◽  
Maria Olho Azul ◽  
Wagner Moreira Lupinacci ◽  
Leandro Machado
Keyword(s):  
Seismic Data ◽  
High Density ◽  
Seismic Facies ◽  
Reservoir Quality ◽  
Framework Analysis ◽  
Carbonate Mounds ◽  
Facies Classification ◽  
Static Modeling ◽  
Well Data ◽  
Low Amplitude

Carbonate mounds, as described herein, often present seismic characteristics such as low amplitude and a high density of faults and fractures, which can easily be oversampled and blur other rock features in simple geobody extraction processes. We have developed a workflow for combining geometric attributes and hybrid spectral decomposition (HSD) to efficiently identify good-quality reservoirs in carbonate mounds within the complex environment of the Brazilian presalt zone. To better identify these reservoirs within the seismic volume of carbonate mounds, we divide our methodology into four stages: seismic data acquisition and processing overview, preconditioning of seismic data using structural-oriented filtering and imaging enhancement, calculation of seismic attributes, and classification of seismic facies. Although coherence and curvature attributes are often used to identify high-density fault and fracture zones, representing one of the most important features of carbonate mounds, HSD is necessary to discriminate low-amplitude carbonate mounds (good reservoir quality) from low-amplitude clay zones (nonreservoir). Finally, we use a multiattribute facies classification to generate a geologically significant outcome and to guide a final geobody extraction that is calibrated by well data and that can be used as a spatial indicator of the distribution of good reservoir quality for static modeling.

Latest developments in seismic texture analysis for subsurface structure, facies, and reservoir characterization: A review

Geophysics ◽  
2011 ◽  
Vol 76 (2) ◽  
pp. W1-W13 ◽  
Author(s):  
Dengliang Gao
Keyword(s):  
Texture Analysis ◽  
Reservoir Characterization ◽  
Preliminary Test ◽  
Seismic Facies ◽  
Least Squares Regression ◽  
Reservoir Properties ◽  
Seismic Amplitude ◽  
Texture Model ◽  
Facies Classification ◽  
Analysis Window

In exploration geology and geophysics, seismic texture is still a developing concept that has not been sufficiently known, although quite a number of different algorithms have been published in the literature. This paper provides a review of the seismic texture concepts and methodologies, focusing on latest developments in seismic amplitude texture analysis, with particular reference to the gray level co-occurrence matrix (GLCM) and the texture model regression (TMR) methods. The GLCM method evaluates spatial arrangements of amplitude samples within an analysis window using a matrix (a two-dimensional histogram) of amplitude co-occurrence. The matrix is then transformed into a suite of texture attributes, such as homogeneity, contrast, and randomness, which provide the basis for seismic facies classification. The TMR method uses a texture model as reference to discriminate among seismic features based on a linear, least-squares regression analysis between the model and the data within an analysis window. By implementing customized texture model schemes, the TMR algorithm has the flexibility to characterize subsurface geology for different purposes. A texture model with a constant phase is effective at enhancing the visibility of seismic structural fabrics, a texture model with a variable phase is helpful for visualizing seismic facies, and a texture model with variable amplitude, frequency, and size is instrumental in calibrating seismic to reservoir properties. Preliminary test case studies in the very recent past have indicated that the latest developments in seismic texture analysis have added to the existing amplitude interpretation theories and methodologies. These and future developments in seismic texture theory and methodologies will hopefully lead to a better understanding of the geologic implications of the seismic texture concept and to an improved geologic interpretation of reflection seismic amplitude.

High-resolution reservoir characterization by an acoustic impedance inversion of a Tertiary deltaic clinoform system in the North Sea

Geophysics ◽  
2010 ◽  
Vol 75 (6) ◽  
pp. O57-O67 ◽  
Author(s):  
Daria Tetyukhina ◽  
Lucas J. van Vliet ◽  
Stefan M. Luthi ◽  
Kees Wapenaar
Keyword(s):  
North Sea ◽  
Seismic Data ◽  
Reservoir Characterization ◽  
Structural Information ◽  
Sampling Rate ◽  
Seismic Inversion ◽  
Acoustic Properties ◽  
The North ◽  
The North Sea ◽  
Well Data

Fluvio-deltaic sedimentary systems are of great interest for explorationists because they can form prolific hydrocarbon plays. However, they are also among the most complex and heterogeneous ones encountered in the subsurface, and potential reservoir units are often close to or below seismic resolution. For seismic inversion, it is therefore important to integrate the seismic data with higher resolution constraints obtained from well logs, whereby not only the acoustic properties are used but also the detailed layering characteristics. We have applied two inversion approaches for poststack, time-migrated seismic data to a clinoform sequence in the North Sea. Both methods are recursive trace-based techniques that use well data as a priori constraints but differ in the way they incorporate structural information. One method uses a discrete layer model from the well that is propagated laterally along the clinoform layers, which are modeled as sigmoids. The second method uses a constant sampling rate from the well data and uses horizontal and vertical regularization parameters for lateral propagation. The first method has a low level of parameterization embedded in a geologic framework and is computationally fast. The second method has a much higher degree of parameterization but is flexible enough to detect deviations in the geologic settings of the reservoir; however, there is no explicit geologic significance and the method is computationally much less efficient. Forward seismic modeling of the two inversion results indicates a good match of both methods with the actual seismic data.

Accuracy of wavelets, seismic inversion, and thin-bed resolution

2017 ◽  
Vol 5 (4) ◽  
pp. T523-T530
Author(s):  
Ehsan Zabihi Naeini ◽  
Mark Sams
Keyword(s):  
Seismic Data ◽  
Reservoir Characterization ◽  
Low Frequency ◽  
Seismic Inversion ◽  
Inversion Method ◽  
Frequency Content ◽  
Frequency Model ◽  
North West ◽  
The North ◽  
Well Data

Broadband reprocessed seismic data from the North West Shelf of Australia were inverted using wavelets estimated with a conventional approach. The inversion method applied was a facies-based inversion, in which the low-frequency model is a product of the inversion process itself, constrained by facies-dependent input trends, the resultant facies distribution, and the match to the seismic. The results identified the presence of a gas reservoir that had recently been confirmed through drilling. The reservoir is thin, with up to 15 ms of maximum thickness. The bandwidth of the seismic data is approximately 5–70 Hz, and the well data used to extract the wavelet used in the inversion are only 400 ms long. As such, there was little control on the lowest frequencies of the wavelet. Different wavelets were subsequently estimated using a variety of new techniques that attempt to address the limitations of short well-log segments and low-frequency seismic. The revised inversion showed greater gas-sand continuity and an extension of the reservoir at one flank. Noise-free synthetic examples indicate that thin-bed delineation can depend on the accuracy of the low-frequency content of the wavelets used for inversion. Underestimation of the low-frequency contents can result in missing thin beds, whereas underestimation of high frequencies can introduce false thin beds. Therefore, it is very important to correctly capture the full frequency content of the seismic data in terms of the amplitude and phase spectra of the estimated wavelets, which subsequently leads to a more accurate thin-bed reservoir characterization through inversion.

Evaluation of uncertainty propagation in seismic facies classification

2014 ◽  
Author(s):  
Cheng Yuan* ◽  
Jingye Li ◽  
Xiaohong Chen ◽  
Fangfang Wang
Keyword(s):  
Uncertainty Propagation ◽  
Seismic Facies ◽  
Facies Classification

A seismic facies classification method based on the convolutional neural network and the probabilistic framework for seismic attributes and spatial classification

2019 ◽  
Vol 7 (3) ◽  
pp. SE225-SE236 ◽  
Author(s):  
Zhege Liu ◽  
Junxing Cao ◽  
Yujia Lu ◽  
Shuna Chen ◽  
Jianli Liu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Early Stage ◽  
Seismic Attributes ◽  
Seismic Facies ◽  
Classification Method ◽  
Probabilistic Framework ◽  
Spatial Classification ◽  
Facies Classification ◽  
Maximal Information Coefficient

In the early stage of oil and gas exploration, due to the lack of available drilling data, the automatic seismic facies classification technology mainly relies on the unsupervised clustering method combined with the seismic multiattribute. However, the clustering results are unstable and have no clear geologic significance. The supervised classification method based on manual interpretation can provide corresponding geologic significance, but there are still some problems such as the discrete classification results and low accuracy. To solve these problems, inspired by hyperspectral and spatial probability distribution technology, we have developed a classification framework called the probabilistic framework for seismic attributes and spatial classification (PFSSC). It can improve the continuity of the classification results by combining the classification probability and the spatial partial probability of the classifier output. In addition, the convolutional neural network (CNN) is a typical classification algorithm in deep learning. By convolution and pooling, we could use it to extract features of complex nonlinear objects for classification. Taking advantage of the combination of PFSSC and CNN, we could effectively solve the existing problems mentioned above in seismic facies classification. It is worth mentioning that we select seismic the multiattribute by maximal information coefficient (MIC) before the seismic facies classification. Finally, using the CNN-PFSSC and MIC methods, we can obtain high accuracy in the test set, reasonable continuity within the same seismic facies, clear boundaries between different seismic facies, and seismic facies classification results consistent with sedimentological laws.

Impact of phase rotation on reservoir characterization and implementation of seismic well tie technique for calibration offshore Nile Delta, Egypt

2020 ◽  
Vol 39 (5) ◽  
pp. 346-352
Author(s):  
Mohamed G. El-Behiry ◽  
Mohamed S. Al Araby ◽  
Ramy Z. Ragab
Keyword(s):  
Seismic Data ◽  
Reservoir Characterization ◽  
Nile Delta ◽  
Seismic Wavelet ◽  
Well Location ◽  
Formation Behavior ◽  
Dynamic Components ◽  
Well Data ◽  
Phase Amplitude ◽  
Seismic Anomalies

Seismic wavelets are dynamic components that result in a seismic trace when convolved with reflectivity series. The seismic wavelet is described by three components: amplitude, frequency, and phase. Amplitude and frequency are considered static because they mainly affect the appearance of a seismic event. Phase can have a large effect on seismic appearance by changing the way it describes the subsurface. Knowing the wavelet properties of certain seismic data facilitates the process of interpretation by providing an understanding of the appearance of regional geologic markers and hydrocarbon-bearing formation behavior. The process through which seismic data wavelets are understood is called seismic well tie. Seismic well tie is the first step in calibrating seismic data in terms of polarity and phase. It ensures that the seismic data are descriptive to regional markers, well markers, and discoveries (if they exist). The step connects well data to seismic data to ensure that the seismic correctly describes well results at the well location. It then extends the understanding of seismic behavior to the rest of the area covered by the seismic data. Good seismic well tie will greatly reduce uncertainties accompanying seismic interpretation. One important outcome of the seismic well tie process is understanding the phase of seismic data, which affects how seismic data will reflect a known geologic marker or hydrocarbon-bearing zone. This understanding can be useful in quantifying discoveries attached to seismic anomalies and extending knowledge from the well location to the rest of the area covered by seismic data.

A comparison of classification techniques for seismic facies recognition

2015 ◽  
Vol 3 (4) ◽  
pp. SAE29-SAE58 ◽  
Author(s):  
Tao Zhao ◽  
Vikram Jayaram ◽  
Atish Roy ◽  
Kurt J. Marfurt
Keyword(s):  
Seismic Data ◽  
Gaussian Mixture ◽  
Seismic Facies ◽  
Support Vector ◽  
3D Seismic ◽  
Generative Topographic Mapping ◽  
Learning Methods ◽  
Facies Classification ◽  
Data Volume ◽  
3D Seismic Data

During the past decade, the size of 3D seismic data volumes and the number of seismic attributes have increased to the extent that it is difficult, if not impossible, for interpreters to examine every seismic line and time slice. To address this problem, several seismic facies classification algorithms including [Formula: see text]-means, self-organizing maps, generative topographic mapping, support vector machines, Gaussian mixture models, and artificial neural networks have been successfully used to extract features of geologic interest from multiple volumes. Although well documented in the literature, the terminology and complexity of these algorithms may bewilder the average seismic interpreter, and few papers have applied these competing methods to the same data volume. We have reviewed six commonly used algorithms and applied them to a single 3D seismic data volume acquired over the Canterbury Basin, offshore New Zealand, where one of the main objectives was to differentiate the architectural elements of a turbidite system. Not surprisingly, the most important parameter in this analysis was the choice of the correct input attributes, which in turn depended on careful pattern recognition by the interpreter. We found that supervised learning methods provided accurate estimates of the desired seismic facies, whereas unsupervised learning methods also highlighted features that might otherwise be overlooked.

Sign in / Sign up

Export Citation Format

Share Document