Improvement of generalization capability of two-dimensional salt segmentation via iterative semi-supervised learning

2022 ◽  
pp. 1-43
Author(s):  
Lingxiao Jia ◽  
Satyakee Sen ◽  
Subhashis Mallick
Keyword(s):  
Oil And Gas ◽  
Three Dimensional ◽  
Ground Truth ◽  
Unlabeled Data ◽  
New Method ◽  
Two Dimensional ◽  
Learning Tools ◽  
Generalization Capability ◽  
Oil And Gas Exploration ◽  
Depth Images

Accurate interpretations of subsurface salts are vital to oil and gas exploration. Manually interpreting them from seismic depth images, however, is labor-intensive. Consequently, use of deep learning tools such as a convolutional neural network for automatic salt interpretation recently became popular. Because of poor generalization capabilities, interpreting salt boundaries using these tools is difficult when labeled data are available from one geological region and we like to make predictions for other nearby regions with varied geological features. At the same time, due to vast amount of the data involved and the associated computational complexities needed for training, such generalization is necessary for solving practical salt interpretation problems. In this work, we propose a semi-supervised training, which allows the predicted model to iteratively improve as more and more information is distilled from the unlabeled data into the model. In addition, by performing mixup between labeled and unlabeled data during training, we encourage the predicted models to linearly behave across training samples; thereby improving the generalization capability of the method. For each iteration, we use the model obtained from previous iteration to generate pseudo labels for the unlabeled data. This automated consecutive data distillation allows our model prediction to improve with iteration, without any need for human intervention. To demonstrate the effectiveness and efficiency, we apply the method on two-dimensional images extracted from a real three-dimensional seismic data volume. By comparing our predictions and fully supervised baseline predictions with those that were manually interpreted and we consider as “ground truth”, we find than the prediction quality our new method surpasses the baseline prediction. We therefore conclude that our new method is a viable tool for automated salt delineation from seismic depth images.

scholarly journals Three-Dimensional Coordination Number from Two-Dimensional Measurements: A New Method

1986 ◽  
Vol 32 (112) ◽  
pp. 391-396 ◽  
Author(s):  
Richard B. Alley
Keyword(s):  
Coordination Number ◽  
Three Dimensional ◽  
New Method ◽  
Two Dimensional ◽  
Shape Factors ◽  
Important Measure ◽  
Dimensional Measurements ◽  
Average Bond

AbstractThe average three-dimensional coordination number, n3, is an important measure of firn structure. The value of n3 can be estimated from n2, the average measured two-dimensional coordination number, and from a function, Γ, that depends only on the ratio of average bond radius to grain radius in the sample. This method is easy to apply and does not require the use of unknown shape factors or tunable parameters.

scholarly journals A new method to estimate point thermal transmittance based on combined two-dimensional heat flow calculation

2020 ◽  
Vol 172 ◽  
pp. 08005
Author(s):  
Jaanus Hallik ◽  
Targo Kalamees
Keyword(s):  
Heat Flow ◽  
Three Dimensional ◽  
Multiple Linear Regression Model ◽  
Building Envelope ◽  
New Method ◽  
Two Dimensional ◽  
Thermal Transmittance ◽  
Thermal Bridge ◽  
Thermal Bridges ◽  
Third Dimension

A well-insulated, airtight and thermal bridge free building envelope is a key factor for nearly zero energy buildings (nZEB). However, increased insulation thickness and minimized air leakages increase the effect of thermal bridges on overall energy efficiency of the nZEBs. Although several more prominent linear thermal bridges are accounted for in the practice the three-dimensional heat flow through vast array of fixation elements, mounting brackets and other point thermal bridges are usually neglected due to time-consuming model preparation routine, lack of input data as well as high number of different thermal bridges that have to be assessed for a single project. In this study a new method was proposed for predicting three-dimensional heat flow and the point thermal transmittance of thermal bridges caused by full or partial penetration of the building envelope with metal elements with uniform geometry in third dimension based on multiple two-dimensional numerical heat flow calculations. A new parameter (equivalent length of thermal bridge) was defined which incorporates the effect of additional thermal transmittance in third dimension when multiplied by the difference of two thermal coupling coefficients derived for two-dimensional cross section. Multiple linear regression model was fitted on database with 102 cases and verified with separate case of window to wall connection incorporating metal penetration at fixation points. The proposed methodology can be useful in general practice where the design team lacks the skills or software tools for conducting detailed numerical analysis in three dimensions.

Beamtrees: Compact Visualization of Large Hierarchies

2003 ◽  
Vol 2 (1) ◽  
pp. 31-39 ◽  
Author(s):  
Frank van Ham ◽  
Jarke J. van Wijk
Keyword(s):  
Hierarchical Structure ◽  
User Study ◽  
Three Dimensional ◽  
New Method ◽  
Data Sets ◽  
Two Dimensional ◽  
Hierarchical Data ◽  
Organization Structures ◽  
The Hierarchical Structure

Beamtrees are a new method for the visualization of large hierarchical data sets, such as directory structures and organization structures. Nodes are shown as stacked circular beams such that both the hierarchical structure as well as the size of nodes are depicted. The dimensions of beams are calculated using a variation of the treemap algorithm. Both a two-dimensional and a three-dimensional variant are presented. A small user study indicated that beamtrees are significantly more effective than nested treemaps and cushion treemaps for the extraction of global hierarchical information.

A Modified Matlock–Duffing Model for Two-Dimensional Ice-Induced Vibrations of Offshore Structures With Geometric Nonlinearities

2015 ◽  
Vol 138 (1) ◽  
Author(s):  
Hugh McQueen ◽  
Narakorn Srinil
Keyword(s):  
Oil And Gas ◽  
Numerical Models ◽  
Wedge Angle ◽  
Offshore Structures ◽  
Two Dimensional ◽  
Geometric Nonlinearities ◽  
Oil And Gas Exploration ◽  
Nonlinear Stress ◽  
Exploration And Production ◽  
Duffing Model

Oil and gas exploration and production have been expanding in Arctic waters. However, numerical models for predicting the ice-induced vibrations (IIV) of offshore structures are still lacking in the literature. This study aims to develop a mathematical reduced-order model for predicting the two-dimensional IIV of offshore structures with geometric coupling and nonlinearities. A cylindrical structure subject to a moving uniform ice sheet is analyzed using the well-known Matlock model, which, in the present study, is extended and modified to account for a new empirical nonlinear stress–strain rate relationship determining the maximum compressive stress (MCS) of the ice. The model is further developed through the incorporation of ice temperature, brine content, air volume, grain size, ice thickness, and ice wedge angle effects on the ice compressive strength. These allow the effect of multiple ice properties on the ice–structure interaction to be investigated. A further advancement is the inclusion of an equation allowing the length of failed ice at a point of failure to vary with time. A mixture of existing equations and newly proposed empirical relationships is used. Structural geometric nonlinearities are incorporated into the numerical model through the use of Duffing oscillators, a technique previously proposed in vortex-induced vibration studies. The model is validated against results from the literature and provides new insights into IIV responses including the quasi-static, randomlike chaotic, and locked-in motions, depending on the ice velocity and system nonlinearities. This numerical Matlock–Duffing model shows a potential to be used in future IIV analysis of Arctic cylindrical structures, particularly fixed offshore structures, such as lighthouses, gravity bases, and wind turbine monopiles.

A new method for underwater dynamic gravimetry based on multisensor integrated navigation

Geophysics ◽  
2020 ◽  
Vol 85 (3) ◽  
pp. G69-G80
Author(s):  
Zhiming Xiong ◽  
Juliang Cao ◽  
Kaixun Liao ◽  
Meiping Wu ◽  
Shaokun Cai ◽  
...  
Keyword(s):  
Oil And Gas ◽  
Finite Impulse Response ◽  
Gravity Data ◽  
Satellite System ◽  
New Method ◽  
Integrated Navigation ◽  
Oil And Gas Exploration ◽  
Half Wavelength ◽  
Marine Gravity ◽  
Global Navigation Satellite

Underwater gravity information plays a major role in deepwater oil and gas exploration. To realize underwater dynamic gravimetry, we have developed a strapdown gravimeter mounted in a pressure capsule for adaption to the underwater environment and we adopted a two-stage towed underwater gravimetry scheme. An improved strapdown gravimeter and other underwater sensors were installed in a towed vessel to form an underwater dynamic gravimetry system. Because the global navigation satellite system cannot be used for underwater dynamic gravimetry, we developed a new method based on underwater multisensor integrated navigation, in which a federal Kalman filter was applied for error estimation. This new method allowed us to obtain the accurate attitude, velocity, and position necessary for gravity estimation. In addition, the gravity data can then be extracted from the noisy data through finite impulse response low-pass filtering. We acquired the underwater gravity data at a depth of 300 m to test the validity of the new method and evaluate the accuracy of the underwater gravity system. The results indicated a repeatability from 0.85 to 0.96 mGal at a half wavelength of approximately 0.2 km and also indicated good consistency with the marine gravity data.

scholarly journals Magnetotelluric regional strike

Geophysics ◽  
1982 ◽  
Vol 47 (6) ◽  
pp. 932-937 ◽  
Author(s):  
T. D. Gamble ◽  
W. M. Goubau ◽  
R. Miracky ◽  
J. Clarke
Keyword(s):  
Baja California ◽  
Weighting Function ◽  
Three Dimensional ◽  
Geothermal Field ◽  
New Method ◽  
Weighted Sums ◽  
Impedance Tensor ◽  
Two Dimensional ◽  
Cerro Prieto

A new method for determining regional strikes from the magnetotelluric impedance tensor Z and tipper T is presented. It involves the minimization of weighted sums of the squared magnitudes of elements of Z or T over all frequencies and all stations of interest. When applied to data from the Mexicali Valley, Baja California around the Cerro Prieto geothermal field for a particular weighting function, the method yielded orientations that agree to within 2.9 degrees for three lines with a total of 16 stations. The consistency of orientations can be attributed in part to the pronounced two‐dimensional (2-D) geologic characteristics of the area, but the techniques also proved stable at stations near three‐dimensional (3-D) inhomogeneities.

Foot shape prediction using elliptical Fourier analysis

2017 ◽  
Vol 88 (9) ◽  
pp. 1026-1037 ◽  
Author(s):  
Ge Wu ◽  
Duan Li ◽  
Pengpeng Hu ◽  
Yueqi Zhong ◽  
Ning Pan
Keyword(s):  
Fourier Analysis ◽  
Principle Component Analysis ◽  
Three Dimensional ◽  
New Method ◽  
Two Dimensional ◽  
Training Set ◽  
Principle Component ◽  
Shape Prediction ◽  
Elliptical Fourier Analysis ◽  
The Relationship

In this paper, a new method was proposed to establish the relationship between three-dimensional (3D) foot shapes and their two-dimensional (2D) foot silhouettes, through which a complete 3D foot shape can be predicted by simply inputting its two 2D silhouettes. 3D foot scans of 80 participants were randomly selected as the training set, and those of another 20 participants were used as the testing set. Elliptical Fourier analysis (EFA) and principle component analysis (PCA) were adopted to parameterize the 3D foot shapes. A linear regressive model was then developed to predict the 3D foot shape with the foot silhouettes. Experiment results indicated individual 3D foot shape can be predicted with a mean error between 1.21 and 1.27 mm, which can provide enough accuracy for the fit evaluation of footwear.

scholarly journals Simulation of Three-Dimensional, Two-Phase Flow In Oil and Gas Reservoirs

1967 ◽  
Vol 7 (04) ◽  
pp. 377-388 ◽  
Author(s):  
K.H. Coats ◽  
R.L. Nielsen ◽  
Mary H. Terhune ◽  
A.G. Weber
Keyword(s):  
Oil And Gas ◽  
Flow Behavior ◽  
Three Dimensional ◽  
Two Dimensional ◽  
Gas Reservoirs ◽  
Three Dimensional Flow ◽  
Cross Sectional ◽  
Dimensional Flow ◽  
Reservoir Flow ◽  
Fluid Phases

COATS, K.H., THE U. OF TEXAS, AUSTIN, TEX. NIELSEN, R.L., ESSO PRODUCTION RESEARCH CO., HOUSTON, TEX. MEMBERS AIME TERHUNE, MARY H., AMERICAN AIRLINES, TULSA, OKLA., WEBER, A.G., ESSO PRODUCTION RESEARCH CO., HOUSTON, TEX. MEMBER AIME Abstract Two computer-oriented techniques for simulating the three-dimensional flow behavior of two fluid phases in petroleum reservoirs were developed. Under the first technique the flow equations are solved to model three-dimensional flow in a reservoir. The second technique was developed for modeling flow in three-dimensional media that have sufficiently high permeability in the vertical direction so that vertical flow is not seriously restricted. Since this latter technique is a modified two-dimensional areal analysis, suitably structured three-dimensional reservoirs can be simulated at considerably lower computational expenses than is required using the three-dimensional analysis. A quantitative criterion is provided for determining when vertical communication is good enough to permit use of the modified two-dimensional areal analysis. The equations solved by both techniques treat both fluids as compressible, and, for gas-oil applications, provide for the evolution of dissolved gas. Accounted for are the effects of relative permeability, capillary pressure and gravity in addition to reservoir geometry and rock heterogeneity. Calculations are compared with laboratory waterflood data to indicate the validity of the analyses. Other results were calculated with both techniques which show the equivalence of the two solutions for reservoirs satisfying the vertical communication criterion. Introduction Obtaining the maximum profits from oil and gas reservoirs during all stages of depletion is the fundamental charge to the reservoir engineering profession. In recent years much quantitative assistance in evaluating field development programs has been goaded by computerized techniques for predicting reservoir flow behavior. Because of the spatially distributed and dynamic nature of producing operations, automatic optimization procedures, such as those now in use for planning refining operations, are not now available for planning reservoir development. However, present mathematical simulation techniques do furnish powerful means for making case studies to help in planning primary recovery operations and in selecting and timing supplemental recovery operations. A number of methods have been reported which simulate the flow of one, two or three fluid phases within porous media of one or two effective spatial dimensions. However, applying computer analyses to actual reservoirs have been limited mostly to two-dimensional areal or cross-sectional flow studies for two immiscible reservoir fluids. To obtain a three-dimensional picture of reservoir performance using such two-dimensional techniques, it has been necessary to interpret the calculations by combining somehow the results from essentially independent areal and cross-sectional studies. To the author's knowledge, the only other three-dimensional computational procedure, in addition to those presented here, was developed by Peaceman and Rachford to simulate the behavior of a laboratory waterflood. Two computational techniques which may be used to simulate three-dimensional flow of two fluid phases are described in this paper. The first method, called the "three-dimensional analysis", employs a fully three-dimensional mathematical model that treats simultaneously both the areal and cross-sectional aspects of reservoir flow. SPEJ P. 377ˆ

scholarly journals Challenges and Prospects of Digital Core-Reconstruction Research

Geofluids ◽  
2019 ◽  
Vol 2019 ◽  
pp. 1-29 ◽  
Author(s):  
Linqi Zhu ◽  
Chong Zhang ◽  
Chaomo Zhang ◽  
Xueqing Zhou ◽  
Zhansong Zhang ◽  
...  
Keyword(s):  
Oil And Gas ◽  
Three Dimensional ◽  
Morphological Characteristics ◽  
Research Area ◽  
Rock Properties ◽  
Reconstruction Method ◽  
Oil And Gas Exploration ◽  
Stochastic Reconstruction ◽  
Reconstruction Methods ◽  
Digital Core

The simulation of various rock properties based on three-dimensional digital cores plays an increasingly important role in oil and gas exploration and development. The accuracy of 3D digital core reconstruction is important for determining rock properties. In this paper, existing 3D digital core-reconstruction methods are divided into two categories: 3D digital cores based on physical experiments and 3D digital core stochastic reconstructions based on two-dimensional (2D) slices. Additionally, 2D slice-based digital core stochastic reconstruction techniques are classified into four types: a stochastic reconstruction method based on 2D slice mathematical-feature statistical constraints, a stochastic reconstruction method based on statistical constraints that are related to 2D slice morphological characteristics, a physics process-based stochastic reconstruction method, and a hybrid stochastic reconstruction method. The progress related to these various stochastic reconstruction methods, the characteristics of constructed 3D digital cores, and the potential of these methods are analysed and discussed in detail. Finally, reasonable prospects are presented based on the current state of this research area. Currently, studies on digital core reconstruction, especially for the 3D digital core stochastic reconstruction method based on 2D slices, are still very rough, and much room for improvement remains. In particular, we emphasize the importance of evaluating functions, multiscale 3D digital cores, multicomponent 3D digital cores, and disciplinary intersection methods in the 3D construction of digital cores. These four directions should provide focus, alongside challenges, for this research area in the future. This review provides important insights into 3D digital core reconstruction.

A Study on the Evolution of Distributary Channels in River-Dominated Deltas Using Numerical Simulations of Hydrodynamic Behaviors

2015 ◽  
Vol 744-746 ◽  
pp. 1050-1055
Author(s):  
Yang Jun Wang ◽  
Tai Ju Yin ◽  
Zhi Hao Deng
Keyword(s):  
Numerical Simulation ◽  
Numerical Model ◽  
Quantitative Methods ◽  
Oil And Gas ◽  
Reference Model ◽  
Three Dimensional ◽  
Hydrodynamic Characteristics ◽  
Oil And Gas Exploration ◽  
Hydrodynamic Behaviors ◽  
Distributary Channel

The Fluvial-dominated delta is one of the extremely important deposition systems in oil and gas exploration. In this paper, the three-dimensional numerical simulation of hydrodynamics has been applied to the precise analysis of the formation of fluvial-dominated deltas and the evolution of their distributary channels. The model has been created using the Delft3D program, and the conditions of the numerical model have been set according to the hydrodynamic characteristics of modern rivers and deltas. The calculation field was 20.5 km in length by 10 km in width. With the Mor-Factor set to 60, the simulation time was 45 days. The formation and the avulsion of the mouth bar, as well as the extension, migration and bifurcation of distributary channels, have been observed and studied through analysis of the simulation results. The vertical cross-section shows that the distributary channel was filled multiple times. According to distributary channel evolution characteristics combined with quantitative methods, the terminal distributary channels can be extremely developed under ideal conditions. Due to the cross-cutting and reform effort of distributary channels, sediments were spread widely and continuously. The results show that the numerical model works well in explaining the process of evolution in fluvial-dominated delta distributary channels. This study not only enables us to quantitatively understand the dynamic processes of terminal distributary channels in fluvial-dominated delta systems, but also provides a reference model for numerical simulation of hydrodynamics in sedimentology study.

Sign in / Sign up

Export Citation Format

Share Document