Elastic FWI for orthorhombic media with lithologic constraints applied via machine learning

Geophysics ◽  
2021 ◽  
pp. 1-86
Author(s):  
Sagar Singh ◽  
Ilya Tsvankin ◽  
Ehsan Zabihi Naeini
Keyword(s):  
Machine Learning ◽  
Model Updating ◽  
Waveform Inversion ◽  
Rock Physics ◽  
P Wave ◽  
Supervised Machine Learning ◽  
Well Logs ◽  
Support Vector ◽  
S Wave ◽  
Trade Offs

Full-waveform inversion (FWI) of 3D wide-azimuth data for elastic orthorhombic media suffers from parameter trade-offs which cannot be overcome without constraining the model-updating procedure. We present an FWI methodology that incorporates geologic constraints to reduce the inversion nonlinearity and increase the resolution of parameter estimation for orthorhombic models. These constraints are obtained from well logs, which can provide rock-physics relationships for different geologic facies. Because the locations of the available well logs are usually sparse, a supervised machine-learning (ML) algorithm (Support Vector Machine) is employed to account for lateral heterogeneity in building the lithologic constraints. The advantages of the facies-based FWI are demonstrated on the modified SEG-EAGE 3D overthrust model, which is made orthorhombic with the symmetry planes that coincide with the Cartesian coordinate planes. We employ a velocity-based parameterization, whose suitability for FWI is studied using the radiation-pattern analysis in a companion paper. Application of the facies-based constraints substantially increases the resolution of the P- and S-wave vertical velocities ( VP0, VS0, and VS1) and, therefore, of the depth scale of the model. Improvements are also observed for the P-wave horizontal and normal-moveout velocities ( VP1, VP2, Vnmo,1, and Vnmo,2) and the S-wave horizontal velocity VS2. However, the velocity Vnmo,3 that depends on Tsvankin’s parameter δ(3) defined in the horizontal plane is not well recovered from the surface data. On the whole, the developed algorithm achieves a much higher spatial resolution compared to unconstrained FWI, even in the absence of recorded frequencies below 2 Hz.

scholarly journals Artificial neural networks as a tool for pattern recognition and electrofacies analysis in Polish palaeozoic shale gas formations

2019 ◽  
Vol 67 (6) ◽  
pp. 1991-2003 ◽  
Author(s):  
Edyta Puskarczyk
Keyword(s):  
Machine Learning ◽  
Pattern Recognition ◽  
Support Vector Machine ◽  
P Wave ◽  
Well Logs ◽  
Support Vector ◽  
Western Slope ◽  
Data Set ◽  
Artificial Neural ◽  
Gas Bearing

Abstract Unconventional oil and gas reservoirs from the lower Palaeozoic basin at the western slope of the East European Craton were taken into account in this study. The aim was to supply and improve standard well logs interpretation based on machine learning methods, especially ANNs. ANNs were used on standard well logging data, e.g. P-wave velocity, density, resistivity, neutron porosity, radioactivity and photoelectric factor. During the calculations, information about lithology or stratigraphy was not taken into account. We apply different methods of classification: cluster analysis, support vector machine and artificial neural network—Kohonen algorithm. We compare the results and analyse obtained electrofacies. Machine learning method–support vector machine SVM was used for classification. For the same data set, SVM algorithm application results were compared to the results of the Kohonen algorithm. The results were very similar. We obtained very good agreement of results. Kohonen algorithm (ANN) was used for pattern recognition and identification of electrofacies. Kohonen algorithm was also used for geological interpretation of well logs data. As a result of Kohonen algorithm application, groups corresponding to the gas-bearing intervals were found. Analysis showed diversification between gas-bearing formations and surrounding beds. It is also shown that internal diversification in gas-saturated beds is present. It is concluded that ANN appeared to be a useful and quick tool for preliminary classification of members and gas-saturated identification.

Predicting reservoir quality in the Bakken Formation, North Dakota, using petrophysics and 3C seismic data

2020 ◽  
Vol 8 (4) ◽  
pp. T851-T868
Author(s):  
Andrea G. Paris ◽  
Robert R. Stewart
Keyword(s):  
North Dakota ◽  
Seismic Data ◽  
Rock Physics ◽  
P Wave ◽  
Well Logs ◽  
Reservoir Quality ◽  
Bakken Formation ◽  
S Wave ◽  
Converted Wave ◽  
Wave Velocities

Combining rock-property analysis with multicomponent seismic imaging can be an effective approach for reservoir quality prediction in the Bakken Formation, North Dakota. The hydrocarbon potential of shale is indicated on well logs by low density, high gamma-ray response, low compressional-wave (P-wave) and shear-wave (S-wave) velocities, and high neutron porosity. We have recognized the shale intervals by cross plotting sonic velocities versus density. Intervals with total organic carbon (TOC) content higher than 10 wt% deviate from lower TOC regions in the density domain and exhibit slightly lower velocities and densities (<2.30 g/cm3). We consider TOC to be the principal factor affecting changes in the density and P- and S-wave velocities in the Bakken shales, where VP/ VS ranges between 1.65 and 1.75. We generate the synthetic seismic data using an anisotropic version of the Zoeppritz equations, including estimated Thomsen’s parameters. For the tops of the Upper and Lower Bakken, the amplitude shows a negative intercept and a positive gradient, which corresponds to an amplitude variation with offset of class IV. The P-impedance error decreases by 14% when incorporating the converted-wave information in the inversion process. A statistical approach using multiattribute analysis and neural networks delimits the zones of interest in terms of P-impedance, density, TOC content, and brittleness. The inverted and predicted results show reasonable correlations with the original well logs. The integration of well log analysis, rock physics, seismic modeling, constrained inversions, and statistical predictions contributes to identifying the areas of highest reservoir quality within the Bakken Formation.

scholarly journals Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques

Materials ◽  
2020 ◽  
Vol 13 (13) ◽  
pp. 2886
Author(s):  
Jong Yil Park ◽  
Sung-Han Sim ◽  
Young Geun Yoon ◽  
Tae Keun Oh
Keyword(s):  
Machine Learning ◽  
Compressive Strength ◽  
Elastic Modulus ◽  
Resonance Frequency ◽  
P Wave ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Regression Equations ◽  
S Wave ◽  
Wave Velocities

The static elastic modulus (Ec) and compressive strength (fc) are critical properties of concrete. When determining Ec and fc, concrete cores are collected and subjected to destructive tests. However, destructive tests require certain test permissions and large sample sizes. Hence, it is preferable to predict Ec using the dynamic elastic modulus (Ed), through nondestructive evaluations. A resonance frequency test performed according to ASTM C215-14 and a pressure wave (P-wave) measurement conducted according to ASTM C597M-16 are typically used to determine Ed. Recently, developments in transducers have enabled the measurement of a shear wave (S-wave) velocities in concrete. Although various equations have been proposed for estimating Ec and fc from Ed, their results deviate from experimental values. Thus, it is necessary to obtain a reliable Ed value for accurately predicting Ec and fc. In this study, Ed values were experimentally obtained from P-wave and S-wave velocities in the longitudinal and transverse modes; Ec and fc values were predicted using these Ed values through four machine learning (ML) methods: support vector machine, artificial neural networks, ensembles, and linear regression. Using ML, the prediction accuracy of Ec and fc was improved by 2.5–5% and 7–9%, respectively, compared with the accuracy obtained using classical or normal-regression equations. By combining ML methods, the accuracy of the predicted Ec and fc was improved by 0.5% and 1.5%, respectively, compared with the optimal single variable results.

scholarly journals Interparameter trade-off quantification for isotropic-elastic full-waveform inversion with various model parameterizations

Geophysics ◽  
2019 ◽  
Vol 84 (2) ◽  
pp. R185-R206 ◽  
Author(s):  
Wenyong Pan ◽  
Kristopher A. Innanen ◽  
Yu Geng ◽  
Junxiao Li
Keyword(s):  
Wave Velocity ◽  
Waveform Inversion ◽  
P Wave ◽  
Physical Parameters ◽  
Full Waveform Inversion ◽  
S Wave ◽  
Trade Off ◽  
Full Waveform ◽  
Trade Offs ◽  
P Wave Velocity

Simultaneous determination of multiple physical parameters using full-waveform inversion (FWI) suffers from interparameter trade-off difficulties. Analyzing the interparameter trade-offs in different model parameterizations of isotropic-elastic FWI, and thus determining the appropriate model parameterization, are critical for efficient inversion and obtaining reliable inverted models. Five different model parameterizations are considered and compared including velocity-density, modulus-density, impedance-density, and two velocity-impedance parameterizations. The scattering radiation patterns are first used for interparameter trade-off analysis. Furthermore, a new framework is developed to evaluate the interparameter trade-off based upon multiparameter Hessian-vector products: Multiparameter point spread functions (MPSFs) and interparameter contamination sensitivity kernels (ICSKs), which provide quantitative, second-order measurements of the interparameter contaminations. In the numerical experiments, the interparameter trade-offs in various model parameterizations are evaluated using the MPSFs and ICSKs. Inversion experiments are carried out with simple Gaussian-anomaly models and a complex Marmousi model. Overall, the parameterization of the P-wave velocity, S-wave velocity, and density, and the parameterization of the P-wave velocity, S-wave velocity, and S-wave impedance perform best for reconstructing all of the physical parameters. Isotropic-elastic FWI of the Hussar low-frequency data set with various model parameterizations verifies our conclusions.

Parameterization analysis and field validation of VTI-elastic full-waveform inversion in a walk-away vertical seismic profile configuration

Geophysics ◽  
2020 ◽  
Vol 85 (3) ◽  
pp. B87-B107 ◽  
Author(s):  
Wenyong Pan ◽  
Kristopher A. Innanen ◽  
Yanfei Wang
Keyword(s):  
Wave Velocity ◽  
Waveform Inversion ◽  
Seismic Profile ◽  
P Wave ◽  
Full Waveform Inversion ◽  
Walk Away ◽  
S Wave ◽  
Full Waveform ◽  
Trade Offs ◽  
Vsp Data

Elastic full-waveform inversion (FWI) in transversely isotropic media with a vertical symmetry axis (VTI) is applied to field walk-away vertical seismic profile (W-VSP) data acquired in Western Canada. The performance of VTI-elastic FWI is significantly influenced by the model parameterization choice. Synthetic analysis based on specific field survey configuration is carried out to evaluate three different VTI-elastic model parameterizations. Interparameter trade-offs are quantified using the recently introduced interparameter contamination sensitivity kernel approach. Synthetic results suggest that neglecting anisotropy leads to inaccurate velocity estimations. For the conventional vertical velocity-Thomsen’s parameter parameterization (i.e., vertical P-wave velocity, vertical S-wave velocity, Thomsen’s parameters [Formula: see text] and [Formula: see text]), a sequential inversion strategy is designed to reduce strong natural interparameter trade-offs. The model parameterizations of elastic-constant ([Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text]) and velocity-based (vertical, horizontal, and normal move-out P-wave velocities and vertical S-wave velocity) models appear to suffer from fewer interparameter trade-offs, providing more reliable velocity and anisotropy models. Results derived from application of VTI-elastic FWI to the field W-VSP data set tend to support the synthetic conclusions. Multiparameter point spread functions are calculated to quantify the local interparameter trade-offs of the inverted models. The output inversion results are interpreted to provide valuable references regarding the target hydrocarbon reservoir.

scholarly journals Predictive Modelling of Employee Turnover in Indian IT Industry Using Machine Learning Techniques

2019 ◽  
Vol 23 (1) ◽  
pp. 12-21 ◽  
Author(s):  
Shikha N. Khera ◽  
Divya
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Confusion Matrix ◽  
Predictive Modelling ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
It Industry ◽  
Knowledge Based ◽  
Employee Attrition

Information technology (IT) industry in India has been facing a systemic issue of high attrition in the past few years, resulting in monetary and knowledge-based loses to the companies. The aim of this research is to develop a model to predict employee attrition and provide the organizations opportunities to address any issue and improve retention. Predictive model was developed based on supervised machine learning algorithm, support vector machine (SVM). Archival employee data (consisting of 22 input features) were collected from Human Resource databases of three IT companies in India, including their employment status (response variable) at the time of collection. Accuracy results from the confusion matrix for the SVM model showed that the model has an accuracy of 85 per cent. Also, results show that the model performs better in predicting who will leave the firm as compared to predicting who will not leave the company.

scholarly journals Supervised Machine Learning Methods and Hyperspectral Imaging Techniques Jointly Applied for Brain Cancer Classification

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3827
Author(s):  
Gemma Urbanos ◽  
Alberto Martín ◽  
Guillermo Vázquez ◽  
Marta Villanueva ◽  
Manuel Villa ◽  
...  
Keyword(s):  
Machine Learning ◽  
Blood Vessel ◽  
Hyperspectral Imaging ◽  
Imaging Techniques ◽  
Venous Blood ◽  
Healthy Tissue ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Arterial Blood

Hyperspectral imaging techniques (HSI) do not require contact with patients and are non-ionizing as well as non-invasive. As a consequence, they have been extensively applied in the medical field. HSI is being combined with machine learning (ML) processes to obtain models to assist in diagnosis. In particular, the combination of these techniques has proven to be a reliable aid in the differentiation of healthy and tumor tissue during brain tumor surgery. ML algorithms such as support vector machine (SVM), random forest (RF) and convolutional neural networks (CNN) are used to make predictions and provide in-vivo visualizations that may assist neurosurgeons in being more precise, hence reducing damages to healthy tissue. In this work, thirteen in-vivo hyperspectral images from twelve different patients with high-grade gliomas (grade III and IV) have been selected to train SVM, RF and CNN classifiers. Five different classes have been defined during the experiments: healthy tissue, tumor, venous blood vessel, arterial blood vessel and dura mater. Overall accuracy (OACC) results vary from 60% to 95% depending on the training conditions. Finally, as far as the contribution of each band to the OACC is concerned, the results obtained in this work are 3.81 times greater than those reported in the literature.

scholarly journals Financial Context News Sentiment Analysis for the Lithuanian Language

2021 ◽  
Vol 11 (10) ◽  
pp. 4443
Author(s):  
Rokas Štrimaitis ◽  
Pavel Stefanovič ◽  
Simona Ramanauskaitė ◽  
Asta Slotkienė
Keyword(s):  
Machine Learning ◽  
Sentiment Analysis ◽  
Short Term Memory ◽  
Machine Learning Algorithms ◽  
Supervised Machine Learning ◽  
Experimental Investigations ◽  
Support Vector ◽  
Applied Machine Learning ◽  
Bayes Algorithm ◽  
Website Content

Financial area analysis is not limited to enterprise performance analysis. It is worth analyzing as wide an area as possible to obtain the full impression of a specific enterprise. News website content is a datum source that expresses the public’s opinion on enterprise operations, status, etc. Therefore, it is worth analyzing the news portal article text. Sentiment analysis in English texts and financial area texts exist, and are accurate, the complexity of Lithuanian language is mostly concentrated on sentiment analysis of comment texts, and does not provide high accuracy. Therefore in this paper, the supervised machine learning model was implemented to assign sentiment analysis on financial context news, gathered from Lithuanian language websites. The analysis was made using three commonly used classification algorithms in the field of sentiment analysis. The hyperparameters optimization using the grid search was performed to discover the best parameters of each classifier. All experimental investigations were made using the newly collected datasets from four Lithuanian news websites. The results of the applied machine learning algorithms show that the highest accuracy is obtained using a non-balanced dataset, via the multinomial Naive Bayes algorithm (71.1%). The other algorithm accuracies were slightly lower: a long short-term memory (71%), and a support vector machine (70.4%).

scholarly journals Optimizing machine learning models for granular NdFeB magnets by very fast simulated annealing

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Hyeon-Kyu Park ◽  
Jae-Hyeok Lee ◽  
Jehyun Lee ◽  
Sang-Koog Kim
Keyword(s):  
Machine Learning ◽  
Simulated Annealing ◽  
Permanent Magnets ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Micromagnetic Simulations ◽  
Ndfeb Magnets ◽  
Average Grain Size ◽  
Macroscopic Properties ◽  
Very Fast Simulated Annealing

AbstractThe macroscopic properties of permanent magnets and the resultant performance required for real implementations are determined by the magnets’ microscopic features. However, earlier micromagnetic simulations and experimental studies required relatively a lot of work to gain any complete and comprehensive understanding of the relationships between magnets’ macroscopic properties and their microstructures. Here, by means of supervised learning, we predict reliable values of coercivity (μ0Hc) and maximum magnetic energy product (BHmax) of granular NdFeB magnets according to their microstructural attributes (e.g. inter-grain decoupling, average grain size, and misalignment of easy axes) based on numerical datasets obtained from micromagnetic simulations. We conducted several tests of a variety of supervised machine learning (ML) models including kernel ridge regression (KRR), support vector regression (SVR), and artificial neural network (ANN) regression. The hyper-parameters of these models were optimized by a very fast simulated annealing (VFSA) algorithm with an adaptive cooling schedule. In our datasets of randomly generated 1,000 polycrystalline NdFeB cuboids with different microstructural attributes, all of the models yielded similar results in predicting both μ0Hc and BHmax. Furthermore, some outliers, which deteriorated the normality of residuals in the prediction of BHmax, were detected and further analyzed. Based on all of our results, we can conclude that our ML approach combined with micromagnetic simulations provides a robust framework for optimal design of microstructures for high-performance NdFeB magnets.

Prediction of CO2 Minimum Miscibility Pressure Using an Augmented Machine-Learning-Based Model

SPE Journal ◽  
2021 ◽  
pp. 1-13
Author(s):  
Utkarsh Sinha ◽  
Birol Dindoruk ◽  
Mohamed Soliman
Keyword(s):  
Machine Learning ◽  
Phase Behavior ◽  
Hybrid Method ◽  
Gas Injection ◽  
Supervised Machine Learning ◽  
Design Parameters ◽  
Support Vector ◽  
Hydrocarbon Gases ◽  
Minimum Miscibility Pressure ◽  
Analytical Correlation

Summary Minimum miscibility pressure (MMP) is one of the key design parameters for gas injection projects. It is a physical parameter that is a measure of local displacement efficiency while subject to some constraints due to its definition. Also, the MMP value is used to tune compositional models along with proper fluid description constrained with other available basic phase behavior data, such as bubble point pressure and volumetric properties. In general, carbon dioxide (CO2) and hydrocarbon gases are the most common gases used for (or screened for) gas injection processes, and because of recent focus, they are used to screen for the coupling of CO2-sequestration and CO2-enhanced oil recovery (EOR) projects. Because the CO2/oil phase behavior is quite different than the hydrocarbon gas/oil phase behavior, researchers developed specialized correlations for CO2 or CO2-rich streams. Therefore, there is a need for a tool with expanded range capabilities for the estimation of MMP for CO2 gas streams. The only known and widely accepted measurement technique for MMP that is coherent with its formal definition is the use of a slimtube apparatus. However, the use of slimtube restricts the amount of data available, even though there are other alternative techniques presented over the last three decades, which all have various limitations (Dindoruk et al. 2021). Due to some of the complexities highlighted in Dindoruk et al. (2021) and time and resource requirements, there have been a number of correlations developed in the literature using mostly classical regression techniques with relatively sparse data using various combinations of limited input data (Cronquist 1978; Lee 1979; Yellig and Metcalfe 1980; Alston et al. 1985; Glaso 1985; Jaubert et al. 1998; Emera and Sarma 2005; Yuan et al. 2005; Ahmadi et al. 2010; Ahmadi and Johns 2011). In this paper, we present two separate approaches for the calculation of the MMP of an oil for CO2 injection: analytical correlation in which the correlation coefficients were tuned using linear support vector machines (SVMs) (Press et al. 2007; MathWorks 2020; RDocumentation 2020b; Cortes and Vapnik 1995) and using a hybrid method (i.e., superlearner model), which consists of the combination of random forest (RF) regression (Breiman 2001) and the proposed analytical correlation. Both models take the compositional analysis of oils up to heptane plus fraction, molecular weight of oil, and the reservoir temperature as input parameters. Based on statistical and data analysis techniques in combination with the help of corresponding crossplots, we showed that the performance of the final proposed method (hybrid method) is superior to all the leading correlations (Cronquist 1978; Lee 1979; Yellig and Metcalfe 1980; Alston et al. 1985; Glaso 1985; Emera and Sarma 2005; Yuan et al. 2005) and supervised machine-learning (Metcalfe 1982) methods considered in the literature (Altman 1992; Chambers and Hastie 1992; Chapelle and Vapnik 2000; Breiman 2001; Press et al. 2007; MathWorks 2020). The proposed model works for the widest spectrum of MMPs from 1,000 to 4,900 psia, which covers the entire range of oils within the scope of CO2 EOR based on the widely used screening criteria (Taber et al. 1997a, 1997b).

Sign in / Sign up

Export Citation Format

Share Document