Elastic Attributes Research Articles

Semi‐supervised deep autoencoder for seismic facies classification

ABSTRACTFacies boundaries are critical for flow performance in a reservoir and are significant for lithofacies identification in well interpretation and reservoir prediction. Facies identification based on supervised machine learning methods usually requires a large amount of labelled data, which are sometimes difficult to obtain. Here, we introduce the deep autoencoder to learn the hidden features and conduct facies classification from elastic attributes. Both labelled and unlabelled data are involved in the training process. Then, we develop a semi‐supervised deep autoencoder by taking the mean of intra‐class and the whole population of facies into account to construct a classification regularization term, thereby improving the classification accuracy and reducing the uncertainty. The new method inherits the profits of deep autoencoder and absorbs the information provided by labelled data. The proposed method performs well and produces promising results when it is used to address problems of reservoir prediction and facies identification. The new method is evaluated on both well and seismic data and compared with the conventional deep autoencoder method, which demonstrates its feasibility and superiority with respect to classification accuracy.

Identifying the drivers of liking for texture attributes of Vietnamese sausage

Vietnamese sausage (cha lua), a pork processing food with a characteristic chewy and crunchy texture, are widely consumed in Vietnam. However, there are not many established researchs on hedonic and sensory attributes for this product, especially about the texture attributes. The goal of this study was to investigate consumer's preference as well as identify the drivers of liking for texture of Vietnamese sausage. Eight samples were prepared with various ratio of lean meat, lard and starch to cover a wide range of different texture of Vietnamese sausage. Sixty eight consumers then evaluated these eight samples, rating texture liking on nine-point scale and answering a checkall- that-apply (CATA) question, which consisted of 16 different texture attributes of Vietnamese sausage. The consumers were also asked to check all the approriate attributes to describe their ideal products. The ANOVA and post –hoc test showed samples which had recipes with high lard ratio (20-25%) and low meat ratio (70%) had a significant higher liking scores, while the samples which had recipe with high meat ratio (90%) had the lowest liking scores. The Cochran's Q test showed that 12 attributes had different choice frequency between samples, suggesting the texture of Vietnamese sausage is complex and diverse. The CA results showed all the samples without stach were associated with large air holes, while samples which high meat ratio (80%) were associated with hardness. The ideal Vietnamese sausage product was associated with fattiness, juiciness, springiness and brittleness. Finally, the results of penalty analysis showed that chewy, firm, and elastic attributes had positive impacts on the liking score for Vietnamese sausage products, while attributes that exhibit heterogeneity, such as large air hole and grainy, significantly reduced the liking score of the product. This suggested the importance of mechanical texture attributes and homogenity on the consumer's preference of the Vietnamese sausage.

Extreme learning machine for multivariate reservoir characterization

Reservoir characterization is one of the most important tasks in oil and gas field exploration and development. Different parameters reflect the relevant information of oil and gas fields from diversified aspects. We design a new reservoir characterization framework by introducing extreme learning machine (ELM) that is one of the state-of-the-art methods in machine learning. It is a single hidden layer feedforward neural (SLFN) network, while the input weight and the bias value of the hidden layer are randomly assigned and kept fixed for simplifying the calculation. Based on ELM, we achieve simultaneous prediction of multiple reservoir parameters (including lithofacies, porosity, shale content and saturation etc.) only through one training step. In order to combat overfitting when the number of hidden nodes is inappropriate or the training samples are inadequate, we extend the method by using biased dropout and dropconnect operations to regularize ELM. We describe the new method in detail and analyze its performance with varying input parameters. It is evaluated on well and seismic datasets by exploiting elastic attributes as training input. Compared with traditional SLFN-based method, ELM-based method uses less computational resources and costs less time on training without losing accuracy. The biased dropout and dropconnect operations further enhance the generalization ability.

Anomalous Flexural Elasticities of Graphene Membranes Unveiled by Manipulating Topology.

Mechanical behavior of atomically thin membranes is governed by bending rigidity and the Gaussian modulus. However, owing to methodological drawbacks, these two parameters have not been investigated sufficiently. We employed atomic force microscopy to demonstrate that the bending rigidity can be extracted from a quadratic relationship of adhesion energy with monolayer curvatures of rolled and unrolled graphene. The tip-induced topological defects revealed the Gaussian modulus; to the best of our knowledge, this is the first study on these parameters. Our study may hold great significance because existing investigations have been performed only on flat graphene. The configurational (strain) energy was evaluated via changes in the surface geometry, with subatomic resolution, by three-dimensional analyses of attractive interatomic forces. The mechanical parameters, evaluated at the hollow sites of the honeycomb lattice, were consistent with the isotropic elastic attributes. The remarkably large negative Gaussian modulus, observed when a single carbon atom was located at the center of the tip-induced bump, revealed attractive interactions between the topological defects and geometric potentials of the Gaussian curvature. Our approach will aid in developing two-dimensional materials and understanding cell biology.

Batch seismic inversion using the iterative ensemble Kalman smoother

An ensemble-based method for seismic inversion to estimate elastic attributes is considered, namely the iterative ensemble Kalman smoother. The main focus of this work is the challenge associated with ensemble-based inversion of seismic waveform data. The amount of seismic data is large and, depending on ensemble size, it cannot be processed in a single batch. Instead a solution strategy of partitioning the data recordings in time windows and processing these sequentially is suggested. This work demonstrates how this partitioning can be done adaptively, with a focus on reliable and efficient estimation. The adaptivity relies on an analysis of the update direction used in the iterative procedure, and an interpretation of contributions from prior and likelihood to this update. The idea is that these must balance; if the prior dominates, the estimation process is inefficient while the estimation is likely to overfit and diverge if data dominates. Two approaches to meet this balance are formulated and evaluated. One is based on an interpretation of eigenvalue distributions and how this enters and affects weighting of prior and likelihood contributions. The other is based on balancing the norm magnitude of prior and likelihood vector components in the update. Only the latter is found to sufficiently regularize the data window. Although no guarantees for avoiding ensemble divergence are provided in the paper, the results of the adaptive procedure indicate that robust estimation performance can be achieved for ensemble-based inversion of seismic waveform data.

A one-step Bayesian inversion framework for 3D reservoir characterization based on a Gaussian mixture model — A Norwegian Sea demonstration

We have developed a one-step approach for Bayesian prediction and uncertainty quantification of lithology/fluid classes, petrophysical properties, and elastic attributes conditional on prestack 3D seismic amplitude-variation-with-offset data. A 3D Markov random field prior model is assumed for the lithology/fluid classes to ensure spatially coupled lithology/fluid class predictions in the lateral and vertical directions. Conditional on the lithology/fluid classes, we consider Gauss-linear petrophysical and rock-physics models including depth trends. Then, the marginal prior models for the petrophysical properties and elastic attributes are multivariate Gaussian mixture models. The likelihood model is assumed to be Gauss-linear to allow for analytic computation. A recursive algorithm that translates the Gibbs formulation of the Markov random field into a set of vertical Markov chains is proposed. This algorithm provides a proposal density in a Markov chain Monte Carlo algorithm such that efficient simulation from the posterior model of interest in three dimensions is feasible. The model is demonstrated on real data from a Norwegian Sea gas reservoir. We evaluate the model at the location of a blind well, and we compare results from the proposed model with results from a set of 1D models in which each vertical trace is inverted independently. At the blind well location, we obtain at most a 60% reduction in the root-mean-square error for the proposed 3D model compared to the model without lateral spatial coupling.

A workflow for multimineralogic carbonate rock physics modelling in a Bayesian framework: Brazilian pre‐salt reservoir case study

ABSTRACTWe develop a workflow to estimate elastic attributes ( and ) of Pre‐Salt carbonate reservoirs using the Xu and Payne rock physics model in a depth‐variant multimineralogic fashion and apply it to one calibration and four extrapolation wells in a field in the Brazilian Santos Basin. Some of the parameters were stochastically simulated in a Bayesian framework to handle the model non‐uniqueness. Nuclear magnetic resonance porosity and mineralogic logs and x‐ray powder diffraction data from the studied wells and literature reference values were used for elastic solid media and dual‐porosity network modelling, bringing geological reasoning to the proposed workflow. Using rock physics templates, we improved our understanding of the reservoir petroelastic trends and devised alternative model parameterizations. Since the model was parameterized in a multimineralogic fashion with depth‐variant mineralogic information (e.g. volumes of carbonate, quartz and clay) as input, we improved elastic attribute estimations within the reservoir, increasing the model predictive power, compared to the common practice of allowing only porosity to vary with depth and representing the matrix properties by a depth‐invariant homogeneous or mixed mineral. We also investigated the workflow robustness and non‐uniqueness by studying its dependence on the choice of calibration well.

Influence of Low-Cost Carrier Attributes towards Airline Choice Decision in Bandung-Surabaya Route

Abstract. The number of Indonesian domestic passengers which increased rapidly has encouraged for the airlines to expand their market share by opening routes in the local region and one of the promising region is Bandung. In early 2019, the Low-Cost Carrier has implemented paid baggage policy. The changes of airline conditions and regulations could affect consumer purchasing decisions. The airline have to redesign its strategies and conduct a strategy adjusment to identify the consumer's needs and wants these days. Based on the problem description, this research aims to identify what significant attributes influence passenger's purchasing decision-making and determine what elastic and inelastic attributes. Data collection using an online questionnaire then spread to 235 respondents and were analyzed with Mixed Logit Model using BIOGEME software. The results shows that airline's departure time is a significant attribute towards passenger's airline choice. Ticket price also plays a key role in selecting an airline for passengers when the ticket price increases, Passenger will consider choosing other alternative airlines that are more affordable. The results could be used by an airline company to attract the passenger by improving its marketing strategy. Keywords: Low-cost carrier, airline choice, airline attribute, mixed logit, demand elasticities

Experimental study and simulation of Natural Fibers Reinforced Polyester Matrix composite

In this study elastic, flexural, effect and hardness attributes of date palm fiber and kenaf rein constrained with polyester mixtures are portrayed unexpectedly. The composite samples are manufactured for various fiber loads by Injection moulding technique. Kenaf filaments having better elasticity contrasted and date palm fiber. Because of that reason, kenaf as base material and date palm fiber as filler. To research the mechanical traits of elastic, flexural, impact and hardness tests were proceeded according to ASTM standard. The mechanical test outcomes uncovered an ordinary pattern of an expansion in malleable, flexural, effect and hardness properties to adding common filaments. Great grip between the normal filaments and the polyester matrix is additionally liable for the successful mechanical ability. This outcome infers that kenaf and date palm fiber have. The results of test works were contrasted and reproduced utilizing ABAQUS software. A 3D model has been set up for recreation of the ductile test composite sample which empowers to comprehend the mechanical strength and endure disappointment of the composite material.

Determination of mechanical attributes of natural fibers reinforced polyester bio composites

In this study elastic, flexural, effect and hardness attributes of date palm fiber and kenaf rein constrained with polyester mixtures are portrayed unexpectedly. The composite samples are manufactured for various fiber loads by Injection moulding technique. Kenaf filaments having better elasticity contrasted and date palm fiber. Because of that reason, kenaf as base material and date palm fiber as filler. To research the mechanical traits of elastic, flexural, impact and hardness tests were proceeded according to ASTM standard. The mechanical test outcomes uncovered an ordinary pattern of an expansion in malleable, flexural, effect and hardness properties to adding common filaments. Great grip between the normal filaments and the polyester matrix is additionally liable for the successful mechanical ability. This outcome infers that kenaf and date palm fiber have.

Comparison of physicochemical properties of jackfruit seed starch with potato and rice starches

ABSTRACT The physicochemical properties of jackfruit seed starch (JSS) were investigated and compared with rice (RS) and potato (PS) starch. The amylose content of JSS (25%) was similar to PS (27%) but higher than RS (17%). The small granules of JSS (7.9 μm) was comparable to RS (9.7 μm), and both starches had A-type crystallinity. In contrast, PS had larger granules (43.4 μm) and displayed B-type crystallinity. The DSC results showed that peak gelatinization temperature of JSS (85.0°C) was significantly higher than RS (66.9°C) and PS (63.7°C), corresponding to higher pasting temperatures and lower swelling power at lower temperatures. There was no significant difference in stability between the starch gels after four freeze-thaw cycles. Formation of JSS gels with significantly higher hardness, chewiness and elastic modulus (G′), but lower fracturability demonstrated JSS ability to form strong gels. This would be beneficial for food applications requiring gelation with firm and elastic textural attributes.

Biocasting of an elastin-like recombinamer and collagen bi-layered model of the tunica adventitia and external elastic lamina of the vascular wall.

The development of techniques for fabricating vascular wall models will foster the development of preventive and therapeutic therapies for treating cardiovascular diseases. However, the physical and biological complexity of vascular tissue represents a major challenge, especially for the design and the production of off-the-shelf biomimetic vascular replicas. Herein, we report the development of a biocasting technique that can be used to replicate the tunica adventitia and the external elastic lamina of the vascular wall. Type I collagen embedded with neonatal human dermal fibroblast (HDFn) and an elastic click cross-linkable, cell-adhesive and protease-sensitive elastin-like recombinamer (ELR) hydrogel were investigated as readily accessible and tunable layers to the envisaged model. Mechanical characterization confirmed that the viscous and elastic attributes predominated in the collagen and ELR layers, respectively. In vitro maturation confirmed that the collagen and ELR provided a favorable environment for the HDFn viability, while histology revealed the wavy and homogenous morphology of the ELR and collagen layer respectively, the cell polarization towards the cell-attachment sites encoded on the ELR, and the enhanced expression of glycosaminoglycan-rich extracellular matrix and differentiation of the embedded HDFn into myofibroblasts. As a complementary assay, 30% by weight of the collagen layer was substituted with the ELR. This model proved the possibility to tune the composition and confirm the versatile character of the technology developed, while revealing no significant differences with respect to the original construct. On-demand modification of the model dimensions, number and composition of the layers, as well as the type and density of the seeded cells, can be further envisioned, thus suggesting that this bi-layered model may be a promising platform for the fabrication of biomimetic vascular wall models.

3D modelling of flow units and petrophysical properties in brazilian presalt carbonate

Rock typing into flow units (FU) is a well-known technique for characterizing flow heterogeneities in reservoirs and producing reliable estimations of petrophysical properties, as porosity and permeability. Several methods that correlate pore-throat size with permeability and porosity in the core and well-log domains are available in the literature, being the flow zone indicator (FZI) method the most common used in both clastic and carbonate reservoirs. Extrapolating the flow units rock typing from the core and well-log scales into the whole reservoir is a major challenge due to the lack of correlation with sedimentological facies and available data scale differences. Most 3D generations of flow units and petrophysical properties are merely a geostatistical procedure, without any spatial data constraints (like seismic attribute). We propose a new approach to create flow units occurrence probability volumes as well as seismic derived porosities and permeabilities that are more robust than the usual estimate of petrophysical properties considering sedimentological facies. Considering the seismic scale characteristics and using the concept of decametre flow units and quantitative seismic interpretation in a Brazilian presalt dataset, we indirectly quantified superimposed small-scale effects that obliterate/generate porosity on a larger scale and established a minimum number of correlatable FUs with elastic seismic attributes. We produced 3D volumes of permeability and porosity that are still capable of obtaining complex reservoir flow characteristics and could be directly considered as variables in lateral interpolation of reservoir parameters, seismic 4D interpretations and seismic-assisted history matching.

A new quantitative model and application for overpressure prediction in carbonate formation

The prediction of pore pressure in subsurface formations is important for the estimation of hydrocarbon accumulation and for safe drilling activities. Typically, pore pressure is estimated from either well logs (e.g., P-wave velocity, resistivity) or seismic data (seismic velocities, acoustic impedance) using empirical relations based on compaction laws. Due to the high density and extremely heterogeneous physical properties, it's not feasible to calculated pore pressure in carbonate rocks with these empirical methods. The result of ultrasonic velocity measurements in carbonate rocks shows that velocity is affected by multiple factors, e.g., stress condition, porosity, rock composition, fluid composition, and other rock properties. It's practicable to predict pore pressure with velocity when other factors are kept constant or evaluated quantitatively.In this paper, a quantitative model to compute the elastic attributes as a function of pressure and petrophysical properties is proposed. The model as a rock physical model is based on the poroelasticity theory and describes the effect on the elastic property of stress conditions. The model is validated using laboratory measurements. A case study for this model application from the northeastern Sichuan Basin, including a set of well logs and borehole measurements is also presented. The pore pressure gradient in carbonate reservoirs is ranging from 0.99 to 2.13 MPa/100 m. The predicted result shows the relative error between model predictions and observations measured while drilling is from 0.60% to 10.22%. The method proposed in this study is effective for pore pressure prediction in carbonate formation.

Creating probabilistic 3D models of lithofluid facies using machine-learning algorithms

Mapping facies variations is a fundamental element in the study of reservoir characteristics. From identifying a pay zone to estimating the reservoir capacity, a hydrocarbon field’s development plan depends to a great extent on a reliable model of lithofacies and fluid content variations throughout the reservoir. The starting point usually is creating 1D facies models based on core samples and drilling reports at each well location. Sparse well locations and the inherent heterogeneity in the reservoir properties make it essential to incorporate the resultant 1D models into a 3D model of facies distribution that includes information about the probability of their occurrence. New techniques using machine learning (ML) to build 3D lithofluid facies (LFF) models can incorporate the prediction of different lithofacies regarding their potential hydrocarbon content, along with the uncertainties of the prediction. We have applied a fuzzy inference system, as an expert-oriented approach, and two separate ML algorithms, to different seismic and elastic attributes to model the LFF classes within the Heidrun oil and gas field. The results, compared with the test wells, show that ML methods could successfully predict the distribution of gas and oil sands within the field, in very good agreement with the known fluid contact intervals. Moreover, the predictions of shale and brine sands vary depending on the method but also are consistent with our knowledge of this field. Comparison between the results confirms the higher reliability of ML methods. Importantly, ML methods provide a better way of investigating and quantifying the uncertainty of the predictions. Implementing ML algorithms in reservoir characterization reduces the risk of drilling unnecessary wells due to false discoveries and can lead to more economical development of hydrocarbon resources.

Decameter-Scale Flow-Unit Classification in Brazilian Presalt Carbonates

Summary Rock typing into flow units (FUs) plays a pivotal role in constructing static and dynamic models of petroleum reservoirs. Decisions made by asset teams mostly depend on predictions of how fluids will percolate through the subsurface during the reservoir life cycle. In carbonate settings, dealing with rock typing is complex and can generate a large quantity of units because of diagenetic processes such as dissolution, cementation, and silicification. Most rock-typing methods in carbonates successfully classify small-scale flow heterogeneity in well resolution but fail when interpolating those facies further away from the 1D domain, because of the lack of correlation between FUs and spatial data. Seismic data can be used to detect large-scale FUs and assist the interpolation of small-scale FUs in 3D reservoir volume, thus producing more-realistic static and dynamic models. We propose a modification of the classical rock-typing methods that use permeability (k) vs. porosity (ϕ) plots and electrical properties, with a data set from the Mero Field, part of the giant Libra Field of presalt carbonate reservoirs in offshore Brazil. From the permeability cumulative S-curve analysis, we established major large-scale FUs that maintain part of the carbonate flow heterogeneity and correlate them with the elastic attributes: P-impedance (PI) and S-impedance (SI). In addition, we established a priori PI and SI correlations with the formation-factor (FF) (F) parameter to categorize large-scale FUs using the F vs. k methodology. With the large-scale FUs mapped in seismic data sets, core-plug-scale FUs can be populated into the 3D static and dynamic models using geostatistics tools, thus creating more-realistic reservoir models and assisting asset teams in the decision-making process.

Bernstein copula-based spatial cosimulation for petrophysical property prediction conditioned to elastic attributes

A new methodology for the simulation of spatially distributed petrophysical properties conditioned by elastic attributes as secondary variables is presented. The method, namely Bernstein copula-based spatial cosimulation (BCSCS), is based on Bernstein copula for the estimation of the joint probability function and simulated annealing for the spatial simulation. The proposed method can model complex non-linear relationships between variables in a fully nonparametric approach. The main advantage is that it does not require linear dependence between variables nor any distribution constraint. This method is first validated in a 1-dimensional case at the well-log scale and it is applied in a 2-dimensional case at the seismic scale to predict effective porosity conditioned to P-impedance in a marine hydrocarbon reservoir located offshore Mexico. The uncertainty quantification analysis shows that the BCSCS method significantly reduces the uncertainty compared to the traditional sequential Gaussian cosimulation method.

Ground instability of sinkhole areas indicated by elastic moduli and seismic attributes

SUMMARY Elastic moduli derived from vertical seismic profiles (VSPs) and 2-D SH-wave reflection seismic profiles are used to characterize mechanical properties of rocks in sinkhole areas. VP and VS were used to calculate the Poisson’s ratio and the dynamic shear modulus. The study shows that 2-D shear wave reflection seismics is suited to depict the heterogeneities of the subsurface induced by subsurface erosion. Low shear wave velocities of ca. 120–350 m s–1 and low shear strength values between 25 and 250 MPa are identified for the subsurface erosion horizon that consists of soluble Permian evapourites and the disturbed overlying deposits. These low values are a result of cavities and fractures induced by dissolution, creating unstable zones. In compliance with the shear modulus the Poisson’s ratio derived from the VSPs shows values of 0.38–0.48 for both the presumed subsurface erosion horizon, and the deposits above. This is a further indicator of reduced underground stability. In the VSPs, anomalies of the shear modulus and the Poisson’s ratio correlate with low electrical resistivities of less than 10 Ωm from borehole logs, indicating high conductivity due to fluid content. Further investigation reveals a conversion of S-to-P wave for the subsurface erosion horizon, which is probably the result of dipping layers and an oriented fracture network. Seismic attribute analysis of the 2-D sections shows strong attenuation of high frequencies and low similarity of adjacent traces, which correlate with the degree of subsurface erosion induced wave disturbance of the underground.

Facies Identification Based on Multikernel Relevance Vector Machine

Facies identification is a powerful means to predict reservoirs. We achieve facies identification using a relevance vector machine (RVM) and develop a facies discriminant method based on a multikernel RVM (MKRVM). An RVM has the same functional form as a support vector machine (SVM) that is widely used in geophysics and shows a promising performance in disposing of small-samples, nonlinear and high-dimensional problems. The RVM inherits these superiorities, and its training is implemented under the Bayesian framework. Thus, it can provide probability information about the classified facies, which is critical to evaluate uncertainty of the result. Besides, the penalty parameter of the RVM does not depend on human experience. Compared with single-kernel learning, multikernel learning (MKL) is more flexible. After mapping the original data into a combined space by MKL, the features can be more accurately expressed in the new space, thereby improving the classification accuracy. Therefore, we introduce the RVM into facies classification and extend it to the MKRVM-based facies identification. The proposed method has advantageous properties such as strong generalization ability and high accuracy. First, we apply the approach to well log facies classification with different input features. Then, it is applied to seismic lithofacies identification with inverted elastic attributes to predict the target reservoirs. All the examples verify the effect and potential of the new method.

Repercussion of pressure on thermodynamic, optoelectronic, thermoelectric and magneto-elastic rectitude of cubic LaFeO3: Quantum DFT perspective

Abstract In this article comprehensive study of structural, electronic, elastic, optical, magnetic and thermoelectric attributes of the cubic LaFeO3 (LFO) perovskite under pressure 0–100 GPa has been considered. The structural stability under pressure is inspected under Goldschmidt tolerance factor, enthalpy of formation and with elastic stability criterion. Our results unveiled that LFO retained its ferromagnetic ductile nature up to 100 GPa pressure. The calculated specific heat CV, magnetic moments and electronic band gap are found to decrease with increasing the induced pressure. The depreciation of these physical parameters is well explained under DOS, contour plots and Fermi surface analysis. Optical properties predict strong candidacy of LFO in optoelectronic and magneto-optical applications. Moreover, half-metallic nature affirms its possible application in pressure dependent spintronic devices. The thermoelectric calculations confirm that LFO compound prefers the hole as charge carriers with the p-type character up to 75 GPa, then switches to the electron carriers under 100 GPa.