scholarly journals Bridging kinetic plasma descriptions and single-fluid models

2020 ◽  
Vol 86 (5) ◽  
Author(s):  
A. Crestetto ◽  
F. Deluzet ◽  
D. Doyen
Keyword(s):  
Numerical Methods ◽  
Low Frequency ◽  
Model Coupling ◽  
Fluid Models ◽  
Maxwell System ◽  
Poisson System ◽  
Microscopic Description ◽  
Frequency Model ◽  
Multi Scale

The purpose of this paper is to bridge kinetic plasma descriptions and low-frequency single-fluid models. More specifically, the asymptotics leading to magnetohydrodynamic regimes starting from the Vlasov–Maxwell system are investigated. The analogy with the derivation, from the Vlasov–Poisson system, of a fluid representation for ions coupled to the Boltzmann relation for electrons is also outlined. The aim is to identify asymptotic parameters explaining the transitions from one microscopic description to a macroscopic low-frequency model. These investigations provide groundwork for the derivation of multi-scale numerical methods, model coupling or physics-based preconditioning.

scholarly journals Non-Intrusive Load Disaggregation Based on a Multi-Scale Attention Residual Network

2020 ◽  
Vol 10 (24) ◽  
pp. 9132
Author(s):  
Liguo Weng ◽  
Xiaodong Zhang ◽  
Junhao Qian ◽  
Min Xia ◽  
Yiqing Xu ◽  
...  
Keyword(s):  
Smart Grids ◽  
Recognition Rate ◽  
Low Frequency ◽  
Attention Mechanism ◽  
Learning Ability ◽  
Residual Network ◽  
Multi Scale ◽  
Energy Disaggregation ◽  
Benchmark Datasets ◽  
Load Disaggregation

Non-intrusive load disaggregation (NILD) is of great significance to the development of smart grids. Current energy disaggregation methods extract features from sequences, and this process easily leads to a loss of load features and difficulties in detecting, resulting in a low recognition rate of low-use electrical appliances. To solve this problem, a non-intrusive sequential energy disaggregation method based on a multi-scale attention residual network is proposed. Multi-scale convolutions are used to learn features, and the attention mechanism is used to enhance the learning ability of load features. The residual learning further improves the performance of the algorithm, avoids network degradation, and improves the precision of load decomposition. The experimental results on two benchmark datasets show that the proposed algorithm has more advantages than the existing algorithms in terms of load disaggregation accuracy and judgments of the on/off state, and the attention mechanism can further improve the disaggregation accuracy of low-frequency electrical appliances.

Turbulent Flows with Drops and Bubbles: What Numerical Simulations Can Tell Us – Freeman Scholar Lecture

2021 ◽  
Author(s):  
Giovanni Soligo ◽  
Alessio Roccon ◽  
Alfredo Soldati
Keyword(s):  
Numerical Methods ◽  
Best Practices ◽  
Numerical Simulations ◽  
Turbulent Flows ◽  
Multiphase Flows ◽  
Simulation Analysis ◽  
Droplet Size Distribution ◽  
Limited Range ◽  
Numerical Resolution ◽  
Multi Scale

Abstract Turbulent flows laden with large, deformable drops or bubbles are ubiquitous in nature and in a number of industrial processes. These flows are characterized by a physics acting at many different scales: from the macroscopic length scale of the problem down to the microscopic molecular scale of the interface. Naturally, the numerical resolution of all the scales of the problem, which span about eight to nine orders of magnitude, is not possible, with the consequence that numerical simulations of turbulent multiphase flows impose challenges and require methods able to capture the multi-scale nature of the flow. In this review, we start by describing the numerical methods commonly employed and discussing their advantages and limitations, and then we focus on the issues arising from the limited range of scales that can be possibly solved. Ultimately, the droplet size distribution, a key result of interest for turbulent multiphase flows, is used as a benchmark to compare the capabilities of the different methods and to discuss the main insights that can be drawn from these simulations. Based on this, we define a series of guidelines and best practices that we believe important in the simulation analysis and in the development of new numerical methods.

Prediction of Petrophysical Properties from Seismic Inversion and Neural Network: A case study

2021 ◽  
Author(s):  
Siddharth Garia ◽  
Arnab Kumar Pal ◽  
Karangat Ravi ◽  
Archana M Nair
Keyword(s):  
Neural Network ◽  
Bulk Density ◽  
Low Frequency ◽  
Seismic Inversion ◽  
Well Log ◽  
Petrophysical Properties ◽  
Seismic Section ◽  
Frequency Model ◽  
Seismic Reflection Data ◽  
Reflection Data

<p>Seismic inversion method is widely used to characterize reservoirs and detect zones of interest, i.e., hydrocarbon-bearing zone in the subsurface by transforming seismic reflection data into quantitative subsurface rock properties. The primary aim of seismic inversion is to transform the 3D seismic section/cube into an acoustic impedance (AI) cube. The integration of this elastic attribute, i.e., AI cube with well log data, can thereafter help to establish correlations between AI and different petrophysical properties. The seismic inversion algorithm interpolates and spatially populates data/parameters of wells to the entire seismic section/cube based on the well log information. The case study presented here uses machine learning-neural network based algorithm to extract the different petrophysical properties such as porosity and bulk density from the seismic data of the Upper Assam basin, India. We analyzed three different stratigraphic  units that are established to be producing zones in this basin.</p><p> AI model is generated from the seismic reflection data with the help of colored inversion operator. Subsequently, low-frequency model is generated from the impedance data extracted from the well log information. To compensate for the band limited nature of the seismic data, this low-frequency model is added to the existing acoustic model. Thereafter, a feed-forward neural network (NN) is trained with AI as input and porosity/bulk density as target, validated with NN generated porosity/bulk density with actual porosity/bulk density from well log data. The trained network is thus tested over the entire region of interest to populate these petrophysical properties.</p><p>Three seismic zones were identified from the seismic section ranging from 681 to 1333 ms, 1528 to 1575 ms and 1771 to 1814 ms. The range of AI, porosity and bulk density were observed to be 1738 to 6000 (g/cc) * (m/s), 26 to 38% and 1.95 to 2.46 g/cc respectively. Studies conducted by researchers in the same basin yielded porosity results in the range of 10-36%. The changes in acoustic impedance, porosity and bulk density may be attributed to the changes in lithology. NN method was prioritized over other traditional statistical methods due to its ability to model any arbitrary dependency (non-linear relationships between input and target values) and also overfitting can be avoided. Hence, the workflow presented here provides an estimation of reservoir properties and is considered useful in predicting petrophysical properties for reservoir characterization, thus helping to estimate reservoir productivity.</p>

scholarly journals Study on the sound absorption behavior of multi-component polyester nonwovens: experimental and numerical methods

2018 ◽  
Vol 89 (16) ◽  
pp. 3342-3361 ◽  
Author(s):  
Tao Yang ◽  
Ferina Saati ◽  
Kirill V Horoshenkov ◽  
Xiaoman Xiong ◽  
Kai Yang ◽  
...  
Keyword(s):  
Numerical Methods ◽  
Absorption Coefficient ◽  
Empirical Model ◽  
Surface Impedance ◽  
Sound Absorption ◽  
Low Frequency ◽  
Accurate Method ◽  
Small Error ◽  
Sound Absorption Coefficient ◽  
Acoustical Properties

This study presents an investigation of the acoustical properties of multi-component polyester nonwovens with experimental and numerical methods. Fifteen types of nonwoven samples made with staple, hollow and bi-component polyester fibers were chosen to carry out this study. The AFD300 AcoustiFlow device was employed to measure airflow resistivity. Several models were grouped in theoretical and empirical model categories and used to predict the airflow resistivity. A simple empirical model based on fiber diameter and fabric bulk density was obtained through the power-fitting method. The difference between measured and predicted airflow resistivity was analyzed. The surface impedance and sound absorption coefficient were determined by using a 45 mm Materiacustica impedance tube. Some widely used impedance models were used to predict the acoustical properties. A comparison between measured and predicted values was carried out to determine the most accurate model for multi-component polyester nonwovens. The results show that one of the Tarnow model provides the closest prediction to the measured value, with an error of 12%. The proposed power-fitted empirical model exhibits a very small error of 6.8%. It is shown that the Delany–Bazley and Miki models can accurately predict surface impedance of multi-component polyester nonwovens, but the Komatsu model is less accurate, especially at the low-frequency range. The results indicate that the Miki model is the most accurate method to predict the sound absorption coefficient, with a mean error of 8.39%.

Poststack impedance inversion considering the diffractive component of the wavefield

Geophysics ◽  
2020 ◽  
Vol 85 (1) ◽  
pp. R11-R28 ◽  
Author(s):  
Kun Xiang ◽  
Evgeny Landa
Keyword(s):  
Specular Reflection ◽  
Low Frequency ◽  
Velocity Model ◽  
Forward Modeling ◽  
Small Scale ◽  
Fracture Detection ◽  
Frequency Model ◽  
New Approach ◽  
Impedance Inversion ◽  
Diffraction Imaging

Seismic diffraction waveform energy contains important information about small-scale subsurface elements, and it is complementary to specular reflection information about subsurface properties. Diffraction imaging has been used for fault, pinchout, and fracture detection. Very little research, however, has been carried out taking diffraction into account in the impedance inversion. Usually, in the standard inversion scheme, the input is the migrated data and the assumption is taken that the diffraction energy is optimally focused. This assumption is true only for a perfectly known velocity model and accurate true amplitude migration algorithm, which are rare in practice. We have developed a new approach for impedance inversion, which takes into account diffractive components of the total wavefield and uses the unmigrated input data. Forward modeling, designed for impedance inversion, includes the classical specular reflection plus asymptotic diffraction modeling schemes. The output model is composed of impedance perturbation and the low-frequency model. The impedance perturbation is estimated using the Bayesian approach and remapped to the migrated domain by the kinematic ray tracing. Our method is demonstrated using synthetic and field data in comparison with the standard inversion. Results indicate that inversion with taking into account diffraction can improve the acoustic impedance prediction in the vicinity of local reflector discontinuities.

scholarly journals ON THE ASYMPTOTIC ANALYSIS OF THE DIRAC–MAXWELL SYSTEM IN THE NONRELATIVISTIC LIMIT

2005 ◽  
Vol 02 (01) ◽  
pp. 129-182 ◽  
Author(s):  
PHILIPPE BECHOUCHE ◽  
NORBERT J. MAUSER ◽  
SIGMUND SELBERG
Keyword(s):  
Convergence Result ◽  
Nonrelativistic Limit ◽  
Dirac Spinor ◽  
Maxwell System ◽  
Poisson System ◽  
Pauli Equation ◽  
Existence Time ◽  
Self Consistent Field ◽  
Klein Gordon ◽  
Well Posed

We study the behavior of solutions of the Dirac–Maxwell system (DM) in the nonrelativistic limit c → ∞, where c is the speed of light. DM is a nonlinear system of PDEs obtained by coupling the Dirac equation for a 4-spinor to the Maxwell equations for the self-consistent field created by the moving charge of the spinor. The limit c → ∞, sometimes also called post-Newtonian, yields a Schrödinger–Poisson system, where the spin and magnetic field no longer appear. We prove that DM is locally well-posed for H1 data (for fixed c), and that as c → ∞ the existence time grows at least as fast as log(c), provided the data are uniformly bounded in H1. Moreover, if the datum for the Dirac spinor converges in H1, then the solution of DM converges, modulo a phase correction, in C([0,T];H1) to a solution of a Schrödinger–Poisson system. Our results also apply to a mixed state formulation of DM, and give also a convergence result for the Pauli equation as the "semi-nonrelativistic" limit. The proof relies on modifications of the bilinear null form estimates of Klainerman and Machedon, and extends our previous work on the nonrelativistic limit of the Klein–Gordon–Maxwell system.

scholarly journals Pansharpening with a Gradient Domain GIF Based on NSST

Electronics ◽  
2019 ◽  
Vol 8 (2) ◽  
pp. 229
Author(s):  
Jiao Jiao ◽  
Lingda Wu
Keyword(s):  
High Frequency ◽  
Low Frequency ◽  
Fusion Rule ◽  
Objective Evaluation ◽  
Data Sets ◽  
Multi Scale ◽  
Guided Image Filter ◽  
Gradient Domain ◽  
Spatial Quality

In order to improve the fusion quality of multispectral (MS) and panchromatic (PAN) images, a pansharpening method with a gradient domain guided image filter (GIF) that is based on non-subsampled shearlet transform (NSST) is proposed. First, multi-scale decomposition of MS and PAN images is performed by NSST. Second, different fusion rules are designed for high- and low-frequency coefficients. A fusion rule that is based on morphological filter-based intensity modulation (MFIM) technology is proposed for the low-frequency coefficients, and the edge refinement is carried out based on a gradient domain GIF to obtain the fused low-frequency coefficients. For the high-frequency coefficients, a fusion rule based on an improved pulse coupled neural network (PCNN) is adopted. The gradient domain GIF optimizes the firing map of the PCNN model, and then the fusion decision map is calculated to guide the fusion of the high-frequency coefficients. Finally, the fused high- and low-frequency coefficients are reconstructed with inverse NSST to obtain the fusion image. The proposed method was tested using the WorldView-2 and QuickBird data sets; the subjective visual effects and objective evaluation demonstrate that the proposed method is superior to the state-of-the-art pansharpening methods, and it can efficiently improve the spatial quality and spectral maintenance.

scholarly journals An introductory guide to fluid models with anisotropic temperatures. Part 2. Kinetic theory, Padé approximants and Landau fluid closures

2019 ◽  
Vol 85 (6) ◽  
Author(s):  
P. Hunana ◽  
A. Tenerani ◽  
G. P. Zank ◽  
M. L. Goldstein ◽  
G. M. Webb ◽  
...  
Keyword(s):  
Kinetic Theory ◽  
Fourth Order ◽  
Low Frequency ◽  
Padé Approximants ◽  
Landau Damping ◽  
Order Moment ◽  
Fluid Models ◽  
Frequency Limit ◽  
Long Wavelength ◽  
Pade Approximants

In Part 2 of our guide to collisionless fluid models, we concentrate on Landau fluid closures. These closures were pioneered by Hammett and Perkins and allow for the rigorous incorporation of collisionless Landau damping into a fluid framework. It is Landau damping that sharply separates traditional fluid models and collisionless kinetic theory, and is the main reason why the usual fluid models do not converge to the kinetic description, even in the long-wavelength low-frequency limit. We start with a brief introduction to kinetic theory, where we discuss in detail the plasma dispersion function $Z(\unicode[STIX]{x1D701})$ , and the associated plasma response function $R(\unicode[STIX]{x1D701})=1+\unicode[STIX]{x1D701}Z(\unicode[STIX]{x1D701})=-Z^{\prime }(\unicode[STIX]{x1D701})/2$ . We then consider a one-dimensional (1-D) (electrostatic) geometry and make a significant effort to map all possible Landau fluid closures that can be constructed at the fourth-order moment level. These closures for parallel moments have general validity from the largest astrophysical scales down to the Debye length, and we verify their validity by considering examples of the (proton and electron) Landau damping of the ion-acoustic mode, and the electron Landau damping of the Langmuir mode. We proceed by considering 1-D closures at higher-order moments than the fourth order, and as was concluded in Part 1, this is not possible without Landau fluid closures. We show that it is possible to reproduce linear Landau damping in the fluid framework to any desired precision, thus showing the convergence of the fluid and collisionless kinetic descriptions. We then consider a 3-D (electromagnetic) geometry in the gyrotropic (long-wavelength low-frequency) limit and map all closures that are available at the fourth-order moment level. In appendix A, we provide comprehensive tables with Padé approximants of $R(\unicode[STIX]{x1D701})$ up to the eighth-pole order, with many given in an analytic form.

Sign in / Sign up

Export Citation Format

Share Document