Seismic Velocity Model Building Using Neural Networks: Training Data Design and Learning Generalization

Data-driven artificial neural networks (ANNs) offer a number of advantages over conventional deterministic methods in a wide range of geophysical problems. For seismic velocity model building, judiciously trained ANNs offer the possibility of estimating high-resolution subsurface velocity models. However, a significant challenge of ANNs is training generalization, which is the ability of an ANN to apply the learning from the training process to test data not previously encountered. In the context of velocity model building, this means learning the relationship between velocity models and the corresponding seismic data from a set of training data, and then using acquired seismic data to accurately estimate unknown velocity models. We ask the following question: what type of velocity model structures need be included in the training process so that the trained ANN can invert seismic data from a different (hypothetical) geological setting? To address this question, we create four sets of training models: geologically inspired and purely geometrical, both with and without background velocity gradients. We find that using geologically inspired training data produce models with well-delineated layer interfaces and fewer intra-layer velocity variations. The absence of a certain geological structure in training models, though, hinders the ANN's ability to recover it in the testing data. We use purely geometric training models consisting of square blocks of varying size to demonstrate the ability of ANNs to recover reasonable approximations of flat, dipping, and curved interfaces. However, the predicted models suffer from intra-layer velocity variations and non-physical artifacts. Overall, the results successfully demonstrate the use of ANNs in recovering accurate velocity model estimates, and highlight the possibility of using such an approach for the generalized seismic velocity inversion problem.

Electroosmosis forces EOF driven boundary layer flow for a non-Newtonian fluid with planktonic microorganism: Darcy Forchheimer model

Purpose The study of the electro-osmotic forces (EOF) in the flow of the boundary layer has been a topic of interest in biomedical engineering and other engineering fields. The purpose of this paper is to develop an innovative mathematical model for electro-osmotic boundary layer flow. This type of fluid flow requires sophisticated mathematical models and numerical simulations. Design/methodology/approach The effect of EOF on the boundary layer Williamson fluid model containing a gyrotactic microorganism through a non-Darcian flow (Forchheimer model) is investigated. The problem is formulated mathematically by a system of non-linear partial differential equations (PDEs). By using suitable transformations, the PDEs system is transformed into a system of non-linear ordinary differential equations subjected to the appropriate boundary conditions. Those equations are solved numerically using the finite difference method. Findings The boundary layer velocity is lower in the case of non-Newtonian fluid when it is compared with that for a Newtonian fluid. The electro-osmotic parameter makes an increase in the velocity of the boundary layer. The boundary layer velocity is lower in the case of non-Darcian fluid when it is compared with Darcian fluid and as the Forchheimer parameter increases the behavior of the velocity becomes more closely. Entropy generation decays speedily far away from the wall and an opposite effect occurs on the Bejan number behavior. Originality/value The present outcomes are enriched to give valuable information for the research scientists in the field of biomedical engineering and other engineering fields. Also, the proposed outcomes are hopefully beneficial for the experimental investigation of the electroosmotic forces on flows with non-Newtonian models and containing a gyrotactic microorganism.

Sensitivity analysis on the impact of geometrical and operational variations on turbine hub cavity modes and practical methods to control them

This paper presents an experimental investigation on the impact of different design and operational variations on the instabilities induced at the hub cavity outlet of a turbine. The experiments were conducted at the “LISA” test facility at ETH Zurich. The axial gap at the 2nd stage hub cavity exit was varied, and also three different flow deflectors were implemented at the cavity exit to control the cavity modes (CMs). Furthermore, the turbine pressure ratio was altered to mimic the off-design condition and study the sensitivity of the CMs to this parameter. Measurements were performed using pneumatic, and Fast Response Aerodynamic Probes (FRAP) at stator and rotor exit. In addition, unsteady pressure transducers were installed at the cavity exit wall to measure the characteristic parameters of the CMs. For the small axial gap, distinct and strong CMs were generated, which actively interacted with stator and rotor hub flow structures. Increasing the gap damped the fluctuations; however, a broader range of frequencies was amplified. The flow deflectors successfully suppressed the CMs by manipulating the shear layer velocity profile and blocking the growing instabilities. Eventually, the increase in the turbine pressure ratio strengthened the CMs and vice versa.

See the Unseen through Target-Oriented Dip-Steered Seismic Resolution

Seismic resolution plays an important role not only in interpretation and reservoir characterization but also in seismic inversion and seismic attributes analysis. The resolution depends on several factors, including seismic frequency bandwidth, dominant frequency, and layer velocity. This paper presents a spectral resolution enhancement approach that is based on Non-stationary Differential Resolution (NSDR) that honors the local structural dip, better preserves amplitude and improves target-oriented seismic interpretation. The proposed technology is applied to both 2D and 3D seismic volumes and findings are compared with the oil industry common spectral enhancement algorithms. We demonstrate the target-oriented dip steering spectral enhancement method on two 3D field datasets and compare the resulting outcome with those obtained by conventional techniques. It is found that thinly layered subsurface geological features with steeply dipping beds are better defined, with artifacts from the conflicting dips removed.

Delineation of S-wave velocity profiles across the Himalayan Frontal Thrust (HFT) using metaheuristic approaches.

<p>The continent-continent collision between the Indian and Asian Plate formed a series of major faults from north to south along the Himalayan belt. Among these Himalayan Frontal Thrust (HFT) is the southernmost and youngest one and is tectonically very active. Any information on the shear wave velocity distribution across the fault is therefore very important. In this study, we have used the Wide Angle Multichannel Analysis of Surface Wave (WAMASW) to estimate the subsurface shear wave velocity profiles across HFT at Pawalgarh in Uttarakhand, India, using widely used stochastic global search Particle Swarm Optimization (PSO) and Grey wolf Optimization (GWO) algorithms. To gain confidence on the accuracy of the inversion results, we first generated an elastic synthetic seismic shot gather with ground rolls by using the forward modelling scheme of SOFI2D for a two-layer velocity depth model overlying a half-space. The generated gather was then processed in MATLAB to generate the experimental dispersion curve using the Phase shift method. We then extracted the fundamental mode for the gather and inverted it using the standard PSO and GWO algorithms and estimated 1D shear wave velocity profile. After getting acceptable results for the synthetic dataset, we then applied the PSO algorithm to generate the 1D S-wave velocity (Vs) profile across the Himalayan Frontal Thrust (HFT). In the study area, the Rayleigh wave phase velocity for the first shot varies from 444 to 743 m/s. We then obtained the 1D shear wave velocity profiles and a jump in Vs is observed across the HFT indicating variation in the sediment stiffness across the fault.</p><p><strong>Keywords: </strong>WAMASW, dispersion, Meta- Heuristic, PSO, GWO, 1D Shear wave velocity</p><p> </p>

Viscosity Ratio Effect on Drop Deformation in the Boundary Layer

In the present study, the motion of a droplet in the boundary layer is investigated numerically. Volume of Fluid method is employed to solve the governing equations. It is found that the presence of the droplet leads to an increase in the pressure inside the boundary layer and on the wall. The results show that the droplets create a vortex on the bottom surface. The friction coefficient increases due to the presence of the droplets and is reduced before and after the droplet due to the formation of a vortex. It is concluded that increasing the viscosity and reducing the density at the same time will not affect the velocity and friction coefficient. It is shown that increasing the radius of the droplet increases the stress and thus decreases the boundary layer velocity. As the Reynolds number increases, the amount of surface friction coefficient decreases. By adding nanoparticles into the pure water, surface friction coefficient increases, especially in the region where the droplet is present.

Numerical Investigation of the Turbulent Wake-Boundary Interaction in a Translational Cascade of Airfoils and Flat Plate

Rotor stator interaction (RSI) is an important phenomenon influencing performances in the pump, turbine, and compressor. In this paper, the correlation-based transition model is used to study the RSI phenomenon between a translational cascade of airfoils and a flat plat. A comparison was made between computational results and experimental results. The computational boundary layer velocity is in reasonable agreement with the experimental velocity. The thickness of boundary layer decreases as the RSI frequency increases and it increases as the fluid flows downstream. The spectral plots of velocity fluctuations at leading edge x/c = 2 under RSI partial flow condition f = 20 Hz and f = 30 Hz are dominated by a narrowband component. RSI frequency mainly affects the turbulence intensity in the freestream region. However, it has little influence on the turbulence intensity of boundary layer near the wall. A secondary vortex is induced by the wake–boundary layer interaction and it leads to the formation of a thickened laminar boundary layer. The negative-vorticity wake also facilitates the formation of a thickened boundary layer while the positive-vorticity wake has a similar effect, like a calmed region which makes the boundary layer thinner.

Validating N-body code chrono for granular DEM simulations in reduced-gravity environments

ABSTRACT The Discrete Element Method (DEM) is frequently used to model complex granular systems and to augment the knowledge that we obtain through theory, experimentation, and real-world observations. Numerical simulations are a particularly powerful tool for studying the regolith-covered surfaces of asteroids, comets, and small moons, where reduced-gravity environments produce ill-defined flow behaviours. In this work, we present a method for validating soft-sphere DEM codes for both terrestrial and small-body granular environments. The open-source code chrono is modified and evaluated first with a series of simple two-body-collision tests, and then, with a set of piling and tumbler tests. In the piling tests, we vary the coefficient of rolling friction to calibrate the simulations against experiments with 1 mm glass beads. Then, we use the friction coefficient to model the flow of 1 mm glass beads in a rotating drum, using a drum configuration from a previous experimental study. We measure the dynamic angle of repose, the flowing layer thickness, and the flowing layer velocity for tests with different particle sizes, contact force models, coefficients of rolling friction, cohesion levels, drum rotation speeds, and gravity levels. The tests show that the same flow patterns can be observed at the Earth and reduced-gravity levels if the drum rotation speed and the gravity level are set according to the dimensionless parameter known as the Froude number. chrono is successfully validated against known flow behaviours at different gravity and cohesion levels, and will be used to study small-body regolith dynamics in future works.