Search

• Suppressing the spatial dispersion of FD modeling by the improved pix2pix algorithm. • Introducing the Sobel operator for robust neural network training. • Numerical tests demonstrating the significant advantages of the proposed method. Seismic finite-difference (FD) modeling inevitably suffers from the spatial dispersion artifacts. In this paper, we propose a new method to suppress the spatial dispersion of seismic FD modeling using the improved pix2pix algorithm by introducing the Sobel operator. The new method involves the following steps. First, a training data set which consists of a small number of wavefield data is generated. The training data set is composed of the high-accuracy wavefield data modeled by the high-order FD method and the low-accuracy wavefield data modeled by the low-order FD method. Second, the entire set of the wavefield data polluted by the spatial dispersion is computed using the low-order FD method. Third, the improved pix2pix neural network is trained and applied to the entire wavefield data set to suppress the spatial dispersion. Numerical tests on different modeling examples demonstrate the effectiveness of the proposed method, which includes the great accuracy in suppressing the spatial dispersion and the high efficiency for the large scale FD modeling. [ABSTRACT FROM AUTHOR]

The continuous interior penalty finite element method (CIP-FEM) has shown promise in reducing pollution errors when numerically simulating the Helmholtz equation with high wave numbers. However, its reliance on structured meshes largely limits its applicability. To address this limitation, this paper presents a novel approach to CIP-FEM on general unstructured meshes. The interior penalty parameters of the CIP-FEM are determined in an offline phase through minimizing the residual obtained by substituting the plane waves into the CIP finite element equation. To enhance accuracy, a quasi-Newton algorithm is employed to correct the penalty parameters and minimize errors of the CIP finite element approximations to the plane waves. The calculated interior penalty parameters can be saved, enabling the online solutions of CIP-FEM to be obtained by reusing these parameters for arbitrary sources during practical computations. Numerical experiments are presented to demonstrate the significant reduction of pollution errors for both linear and higher-order CIP-FEMs on unstructured meshes. Furthermore, the proposed algorithm successfully simulates high-frequency acoustic fields in a three-dimensional vehicle cabin with greatly reduced errors within certain fixed computational time, demonstrating its practical effectiveness in real-world scenarios. • The optimized CIP-FEM can be applied to unstructured meshes, addressing a notable limitation of standard CIP-FEM. • The penalty parameters are determined in an offline phase through the residual/error-minimization approaches. • The proposed algorithm successfully simulates high-frequency acoustic fields in vehicle cabin in real-world scenarios. [ABSTRACT FROM AUTHOR]

In this paper, we propose an iterative convolution-thresholding method (ICTM) based on prediction-correction for solving the topology optimization problem in steady-state heat transfer equations. The problem is formulated as a constrained minimization problem of the complementary energy, incorporating a perimeter/surface-area regularization term, while satisfying a steady-state heat transfer equation. The decision variables of the optimization problem represent the domains of different materials and are represented by indicator functions. The perimeter/surface-area term of the domain is approximated using Gaussian kernel convolution with indicator functions. In each iteration, the indicator function is updated using a prediction-correction approach. The prediction step is based on the variation of the objective functional by imposing the constraints, while the correction step ensures the monotonically decreasing behavior of the objective functional. Numerical results demonstrate the efficiency and robustness of our proposed method, particularly when compared to classical approaches based on the ICTM. • Decision variables for the topology optimization problem are represented by indicator functions. • The objective functional is approximated using Gaussian kernel convolution with indicator functions. • The prediction step is based on the variation of the objective functional by imposing the constraints. • The correction step ensures the monotonically decreasing behavior of the objective functional. • Numerical results demonstrate the efficiency and robustness of our proposed method. [ABSTRACT FROM AUTHOR]

Inverse problems, i.e., estimating parameters of physical models from experimental data, are ubiquitous in science and engineering. The Bayesian formulation is the gold standard because it alleviates ill-posedness issues and quantifies epistemic uncertainty. Since analytical posteriors are not typically available, one resorts to Markov chain Monte Carlo sampling or approximate variational inference. However, inference needs to be rerun from scratch for each new set of data. This drawback limits the applicability of the Bayesian formulation to real-time settings, e.g., health monitoring of engineered systems, and medical diagnosis. The objective of this paper is to develop a methodology that enables real-time inference by learning the Bayesian inverse map, i.e., the map from data to posteriors. Our approach is as follows. We parameterize the posterior distribution as a function of data. This work outlines two distinct approaches to do this. The first method involves parameterizing the posterior using an amortized full-rank Gaussian guide, implemented through neural networks. The second method utilizes a Conditional Normalizing Flow guide, employing conditional invertible neural networks for cases where the target posterior is arbitrarily complex. In both approaches, we learn the network parameters by amortized variational inference which involves maximizing the expectation of evidence lower bound over all possible datasets compatible with the model. We demonstrate our approach by solving a set of benchmark problems from science and engineering. Our results show that the posterior estimates of our approach are in agreement with the corresponding ground truth obtained by Markov chain Monte Carlo. Once trained, our approach provides the posterior distribution for a given observation just at the cost of a forward pass of the neural network. • An approach eliminating the need to re-run posterior inference from scratch for each new data. • Posterior parameterization using amortized posteriors represented by neural nets. • Formulation of an amortized loss function and training the network using it. • Application to solving benchmark problems in science and engineering. • Posteriors are estimated just at the cost of forward pass of the network once trained. [ABSTRACT FROM AUTHOR]

In this paper, a topology optimization method for rarefied gas flows is proposed. We use the discrete velocity method (DVM) to solve the Boltzmann equation, and construct an extension of the standard DVM algorithm in order to model gas flow in a domain of mixed materials. The proposed extension consists of a correction in convection fluxes as well as an extra source term which mimics reflection boundary conditions, so that the presence of fluid or solid can be characterized by a pseudo density. To evaluate the change in objective functional due to the design variable, an adjoint problem is derived from the steady state condition of the extended DVM, which forms a linear system of equations that can be solved using iterative methods like in standard DVM. Using the design sensitivity obtained from the adjoint system, optimization of flow structures can be achieved. Validity of the proposed method is demonstrated through several numerical examples. [ABSTRACT FROM AUTHOR]

In this paper, we propose a reduced-dimensional smoothed particle hydrodynamics (SPH) formulation for quasi-static and dynamic analyses of plate and shell structures undergoing finite deformation and large rotation. By exploiting Uflyand–Mindlin plate theory, the present surface-particle formulation is able to resolve the thin structures by using only one layer of particles at the mid-surface. To resolve the geometric non-linearity and capture finite deformation and large rotation, two reduced-dimensional linear-reproducing correction matrices are introduced, and weighted non-singularity conversions between the rotation angle and pseudo normal are formulated. A new non-isotropic Kelvin-Voigt damping is proposed especially for the both thin and moderately thick plate and shell structures to increase the numerical stability. In addition, a shear-scaled momentum-conserving hourglass control algorithm with an adaptive limiter is introduced to suppress the mismatches between the particle position and pseudo normal and those estimated with the deformation gradient. A comprehensive set of test problems, for which the analytical or numerical results from literature or those of the volume-particle SPH model are available for quantitative and qualitative comparison, are examined to demonstrate the accuracy and stability of the present method. • Two reduced-dimensional linear-reproducing correction matrices. • Weighted non-singularity conversions between the rotation angle and pseudo normal. • Non-isotropic Kelvin-Voigt damping. • A shear-scaled momentum-conserving hourglass control algorithm. [ABSTRACT FROM AUTHOR]

More than two decades ago, the numerical instabilities of particle clustering and non-physical fractures encountered in the simulations of typical elastic dynamics problems using updated Lagrangian smoothed particle hydrodynamics (ULSPH) was identified as tensile instability. Despite continuous efforts in the past, a satisfactory resolution for the simulations of these problems has remained elusive. In this paper, the concept of hourglass modes, other than tensile instability, is first explored for the discretization of shear force, arguing that the former may actually lead to the numerical instabilities in these simulations. Based on such concept, we present an essentially non-hourglass formulation by utilizing the Laplacian operator which is widely used in fluid simulations. Together with the dual-criteria time stepping, adopted into the simulation of solids for the first time to significantly enhance computational efficiency, a comprehensive set of challenging benchmark cases is used to showcase that our method achieves accurate and stable SPH elastic dynamics. • Longstanding numerical instability issue in SPH solid dynamics clarified. • Essentially non-hourglass formulation presented. • Comprehensive set of challenging benchmark cases given. • Dual-criteria time stepping in SPH solid simulation for the first time. [ABSTRACT FROM AUTHOR]

This paper presents a novel spatial discretisation method for reliable and efficient simulation of Bose-Einstein condensates modelled by the Gross-Pitaevskii equation and the corresponding nonlinear eigenvector problem. The method combines the high-accuracy properties of numerical homogenisation methods with a novel super-localisation approach for the calculation of the basis functions. A rigorous numerical analysis of the nonlinear eigenvector problem demonstrates superconvergence of the ideal approach compared to classical polynomial and multiscale finite element methods, even in low regularity regimes. Numerical tests show that the super-localised method is competitive with spectral methods, particularly in capturing critical physical effects in extreme conditions, such as vortex lattice formation in fast-rotating potential traps. The method's potential is further highlighted through a dynamic simulation of a phase transition from Mott insulator to Bose-Einstein condensate, emphasising its capability for reliable exploration of physical phenomena. • Efficient and reliable calculation of ground and excited states and dynamics of Bose-Einstein condensates. • New energy minimisation method in low regularity regimes using the superlocalised orthogonal decomposition in space. • Proof of optimal convergence when solving the nonlinear eigenvector problem in presence of L 2 potentials. • Numerical comparisons with popular and powerful spectral methods demonstrate the competitiveness of our method. • A laptop simulation capturing the physical behavior of a Bose-Einstein condensate released from an optical lattice in 3d. [ABSTRACT FROM AUTHOR]

In this paper, the new distinct-positivity-preserving (DPP) methods are proposed for fifth-order finite volume multi-resolution WENO schemes with adaptive linear weights (which are termed as the ALW-MR-WENO schemes) [J. Comput. Phys., 493 (2023), 112471] for solving some extreme problems on structured meshes. Associated WENO spatial reconstruction procedures only require two hierarchical unequal-sized central spatial stencils for achieving fifth-order accuracy in smooth regions and keeping the essentially non-oscillatory property in non-smooth regions. One redefines five cell averages after performing such spatial reconstructions and then designs one quartic polynomial and two cubic reconstruction polynomials based on them in one dimension. After that, a new detective process is used to examine the positivity of the point values of such cubic reconstruction polynomials at some checking points inside the target cell. If the negativity happens, a new over-determined compression limiter is carried out to ensure that the compressed polynomials can achieve the positivity in the whole target cell instead of only at some discrete points inside it. The quartic polynomial could use its four quadrature point values by using a four-point Gauss-Lobatto quadrature formula when performing numerical integrations. Since it is easy to rewrite the point values of one quartic polynomial with the point values of two cubic polynomials timing with associated linear weights, both cubic polynomials could use a four-point Gauss-Lobatto quadrature formula or a three-point Simpson's quadrature formula without losing any algebra precisions. By doing this novelty switching methodology between two quadrature formulas, it is the first time to theoretically prove and increase a sufficient CFL number from 1/12 to 1/7.2 for the fifth-order WENO schemes in one dimension. This methodology can be easily extended to multi-dimensions. Finally, the novelty DPP methods for fifth-order ALW-MR-WENO schemes with a larger practical CFL number of 0.6, instead of the sufficient CFL number of 1/7.2, are also simple and robust enough for some extreme problems without timely halving the CFL number to avoid the appearance of negative density and negative pressure. [ABSTRACT FROM AUTHOR]

In this paper, a high order well-balanced asymptotic preserving scheme is presented for the two-dimensional nonlinear shallow water equations over variable bottom topography in all Froude number regimes. To obtain the well-balanced property, the system is first reformulated as a new form by introducing an auxiliary parameter. The flux is then split into a linear stiff part to be treated implicitly and a nonlinear non-stiff part to be treated explicitly, and the source term is treated explicitly. An implicit-explicit Runge-Kutta discontinuous Galerkin scheme is designed for solving the equations. The proposed scheme can be proved to be well-balanced, asymptotic preserving and asymptotically accurate. Finally, several numerical tests are carried out to validate the performance of our proposed scheme. • A new splitting scheme is presented for the 2D nonlinear shallow water equations over variable bottom. • A high order IXEM RKDG method is designed to solve the equations. • The method is proved to be well-balanced, asymptotic preserving and asymptotically accurate. [ABSTRACT FROM AUTHOR]

This paper presents a physics and data co-driven surrogate modeling method for efficient rare event simulation of civil and mechanical systems with high-dimensional input uncertainties. The method fuses interpretable low-fidelity physical models with data-driven error corrections. The hypothesis is that a well-designed and well-trained simplified physical model can preserve salient features of the original model, while data-fitting techniques can fill the remaining gaps between the surrogate and original model predictions. The coupled physics-data-driven surrogate model is adaptively trained using active learning, aiming to achieve a high correlation and small bias between the surrogate and original model responses in the critical parametric region of a rare event. A final importance sampling step is introduced to correct the surrogate model-based probability estimations. Static and dynamic problems with input uncertainties modeled by random field and stochastic process are studied to demonstrate the proposed method. • Physics-data-driven surrogate modeling method for estimating small probabilities. • Fusion of physical model, error correction, active learning, and importance sampling. • Highly efficient for rare event simulation with high-dimensional input uncertainties. [ABSTRACT FROM AUTHOR]

This paper introduces a formulation of the variable density incompressible Navier-Stokes equations by modifying the nonlinear terms in a consistent way. For Galerkin discretizations, the formulation leads to favorable discrete conservation properties without the divergence-free constraint being strongly enforced. In addition, the formulation is shown to make the density field invariant to global shifts. The effect of viscous regularizations on conservation properties is also investigated. Numerical tests validate the theory developed in this work. The new formulation shows superior performance compared to other formulations from the literature, both in terms of accuracy for smooth problems and in terms of robustness. • A new formulation for variable density incompressible flow is introduced. • The formulation leads to improved conservation properties without the divergence-free being strongly enforced. • The effect of viscous regularization on conservation properties is investigated. • Conservation of mass, kinetic energy, squared density, momentum and angular momentum are analyzed. • Mass is conserved by matching or reducing the polynomial degree of density to that of pressure. [ABSTRACT FROM AUTHOR]

Based upon two overlapped, body-unfitted meshes, a type of unified-field monolithic fictitious domain-finite element method (UFMFD-FEM) is developed in this paper for moving interface problems of dynamic fluid-structure interactions (FSI), which is accompanied with high-contrast physical coefficients across the interface and with contacting collisions between the structure and fluidic channel wall in the immersed FSI case. In particular, the proposed novel numerical method consists of a monolithic, stabilized mixed finite element method within the frame of fictitious domain/immersed boundary method (FD/IBM) for generic fluid-structure-contact interaction (FSCI) problems in the Eulerian–updated Lagrangian description, while involving the no-slip type of interface conditions on the fluid-structure interface, and the repulsive contact force/stress on the structural surface when the immersed structure contacts the fluidic channel wall. The developed UFMFD-FEM for FSI/FSCI problems can deal with the structural motion with large rotational and translational displacements and/or large deformation in an accurate and efficient fashion, which is first validated by two benchmark FSI problems and one FSCI model problem, then by experimental results of a realistic FSCI scenario – the microfluidic deterministic lateral displacement (DLD) problem that is applied to isolate circulating tumor cells (CTCs) from blood cells in the blood fluid through a cascaded filter DLD microchip in practice, where a particulate fluid with the pillar obstacles effect in the fluidic channel, i.e., the effects of fluid-structure interaction and structure collision, play significant roles to sort particles (cells) of different sizes with tilted pillar arrays. The developed unified-field, monolithic fictitious domain-based mixed finite element method can be seamlessly extended to more sophisticated, high dimensional FSCI problems with contacting collisions between the moving elastic structure and fluidic channel wall. • A fluid-structure-contact interaction (FSCI) model involving the interaction and collision effects is developed. • A mixed FEM in the monolithic fictitious domain frame is developed with respect to a unified variable pair of velocity and pressure. • Advanced algorithms are designed to handle structural collision, nonlinear system and interpolation between fictitious fluid and structure. • Numerical approaches are validated by benchmark problems, self- and physically designed DLD problems. • Numerical simulations carried out by the developed UFMFD-FEM can help to optimize the design of DLD microchips. [ABSTRACT FROM AUTHOR]

In this paper, we propose and study a numerical scheme for the Navier-Stokes-Fourier system derived and studied by the authors in [12]. This system models hydrostatic free surface flows with density variations depending on temperature or salinity. We show that the finite volume/finite element scheme presented – based on a layer averaged formulation of the model – is well-balanced with regards to the steady state of the lake at rest and preserves the nonnegativity of the water height. A maximum principle on the density is also proved as well as a discrete entropy inequality (when the thermal and viscous effects are neglected). Some numerical validations are finally shown with comparisons to 3D analytical solutions and experiments. [ABSTRACT FROM AUTHOR]

The efficient simulation of fluid-structure interactions at zero Reynolds number requires the use of fast summation techniques in order to rapidly compute the long-ranged hydrodynamic interactions between the structures. One approach for periodic domains involves utilising a compact or exponentially decaying kernel function to spread the force on the structure to a regular grid where the resulting flow and interactions can be computed efficiently using an FFT-based solver. A limitation to this approach is that the grid spacing must be chosen to resolve the kernel and thus, these methods can become inefficient when the separation between the structures is large compared to the kernel width. In this paper, we address this issue for the force-coupling method (FCM) by introducing a modified kernel that can be resolved on a much coarser grid, and subsequently correcting the resulting interactions in a pairwise fashion. The modified kernel is constructed to ensure rapid convergence to the exact hydrodynamic interactions and a positive-splitting of the associated mobility matrix. We provide a detailed computational study of the methodology and establish the optimal choice of the modified kernel width, which we show plays a similar role to the splitting parameter in Ewald summation. Finally, we perform example simulations of rod sedimentation and active filament coordination to demonstrate the performance of fast FCM in application. • A modified kernel and velocity correction scheme are developed to accelerate the force-coupling method (FCM). • In several cases, fast FCM is an order of magnitude faster than the standard FCM algorithm. • Fast FCM is extended to include point torques on the particles. • Fast FCM is shown to yield positive splitting, which is important for simulations involving thermal fluctuations. • The effectiveness of fast FCM is demonstrated through simulations of rod suspensions and arrays of flexible filaments. [ABSTRACT FROM AUTHOR]

Sparse-grid methods have recently gained interest in reducing the computational cost of solving high-dimensional kinetic equations. In this paper, we construct adaptive and hybrid sparse-grid methods for the Vlasov–Poisson–Lenard–Bernstein (VPLB) model. This model has applications to plasma physics and is simulated in two reduced geometries: a 0 x 3 v space homogeneous geometry and a 1 x 3 v slab geometry. We use the discontinuous Galerkin (DG) method as a base discretization due to its high-order accuracy and ability to preserve important structural properties of partial differential equations. We utilize a multiwavelet basis expansion to determine the sparse-grid basis and the adaptive mesh criteria. We analyze the proposed sparse-grid methods on a suite of three test problems by computing the savings afforded by sparse-grids in comparison to standard solutions of the DG method. The results are obtained using the adaptive sparse-grid discretization library ASGarD. [ABSTRACT FROM AUTHOR]

An algorithm is presented to calculate the field components of the magnetic field [ B x (x , y , z) , B y (x , y , z) , B z (x , y , z) ] at a point (x , y , z) in space, from the knowledge of the components [ B x (x , y = 0 , z) , B y (x , y = 0 , z) , B z (x , y = 0 , z) ] on a "reference plane", which is normal to the y -axis at y = 0. The algorithm, which is a general one and is not restricted to fields with mid-plane symmetry is based on the Maclaurin series expansion of the magnetic field components at any point in space in terms of the distance (y) of the point from the reference plane. The coefficients of the Maclaurin series expansion are expressed in terms of the on-plane field components and their partial derivatives with respect to spatial coordinates (x , z). The field components are usually generated from magnetic field measurements on a rectangular grid on the plane. This algorithm was employed in 1986 in the RAYTRACE computer code to help calculate the optical properties of magnets and of the Alternating Gradient Synchrotron (AGS) at the Brookhaven National Laboratory (BNL). A general mathematical formulation of this algorithm based on the differential algebraic method was presented by Makino in 2011. This paper presents the step by step derivation of the algorithm and provides the necessary formulas to be introduced by the reader in any computer code which requires the field components generated by magnetic devices. In addition provides an example of the use of the algorithm and its limitations as applied to a Halbach type magnet with or without median plane symmetry. • The field of accelerator physics relays on magnetic field measurements of the devices used in beam lines or accelerators. • Such magnetic field measurements are limited in space for any particular magnet therefor algorithms like the one presented in this paper are important to calculate the values of the magnetic fields to a larger region of space than the space covered by the measured magnetic fields. [ABSTRACT FROM AUTHOR]

In the present paper, a multi-step reconstruction procedure is proposed for high order finite volume schemes on unstructured grids using compact stencil. The procedure is a recursive algorithm that can eventually provide sufficient relations for high order reconstruction in a multi-step procedure. Two key elements of this procedure are the partial inversion technique and the continuation technique. The partial inversion can be used not only to obtain lower order reconstruction based on existing reconstruction relations, but also to regularize the existing reconstruction relations to provide new relations for higher order reconstructions. The continuation technique is to extend the regularized relations on the face-neighboring cells to current cell as additional reconstruction relations. This multi-step procedure is operationally compact since in each step only the relations defined on a compact stencil are used. In the present paper, the third and fourth order finite volume schemes based on two-step quadratic and three-step cubic reconstructions are studied. [ABSTRACT FROM AUTHOR]

In a previous paper, [1] , a new formulation for isothermal gas flow in a single pipeline including gravity and friction effects has been introduced and solved by finite element methods. A performance analysis has been done by using some tests both manufactured and experimental. The goal of the present work is to extend the ideas to model gas flow in a transportation network. As a first step, in this paper the network only includes pipes connected at nodes which can be emission or consumption points, or simply structural connections between pipes. In addition to the mass and momentum conservation equations inside the pipes that are partial differential equations, we need to write the mass conservation equations at the nodes. The latter appear to be a set of constraints (one per node) whose respective Lagrange multipliers are the (continuous) pressures at the nodes. Thus, the whole model is a Partial Differential-Algebraic Equation (PDAE). For numerical solution, an implicit time discretization is proposed first and then a weak formulation of the resulting semi-discrete problem is introduced. After doing a finite element discretization we are led to solve a nonlinear system of numerical equations per time step what is done by Newton's method. In order to reduce the computing time, the algorithm can be parallelized in such a way that, at each time step, every pipe can be solved separately in one processor (or core). Moreover, we show that the proposed scheme exactly conserves the mass at the nodes and in the whole network, and a new simple and compact way of computing the network line-pack over the time is given. Finally, the introduced methodology is validated by applying the computer program to a triangular network, which is a standard test in references on the subject, and to a real network for which we have got measurements that enable us to create the boundary conditions for the model and also to compare the numerical results with experimental data. • Numerical solution of Euler-like models with friction and variable topography. • An equivalent formulation with one nonlinear PDE of second order in time and space. • An implicit time discretization combined with a finite element method is used. • Accuracy has been tested with experimental data and numerical results from papers. [ABSTRACT FROM AUTHOR]

• Introduce a new third-order conservative sign-preserving and steady-state-preserving RK method. • The ODE solver is A (π 4) -stable. • The technique can be applied to stiff detonation with large time step. In this paper, we develop third-order conservative sign-preserving and steady-state-preserving time integrations and seek their applications in multispecies and multireaction chemical reactive flows. In this problem, the density and pressure are nonnegative, and the mass fraction for the i th species, denoted as z i , 1 ≤ i ≤ M , should be between 0 and 1, where M is the total number of species. There are four main difficulties in constructing high-order bound-preserving techniques for multispecies and multireaction detonations. First of all, most of the bound-preserving techniques available are based on Euler forward time integration. Therefore, for problems with stiff source, the time step will be significantly limited. Secondly, the mass fraction does not satisfy a maximum-principle and hence it is not easy to preserve the upper bound 1. Thirdly, in most of the previous works for gaseous denotation, the algorithm relies on second-order Strang splitting methods where the flux and stiff source terms can be solved separately, and the extension to high-order time discretization seems to be complicated. Finally, most of the previous ODE solvers for stiff problems cannot preserve the total mass and the positivity of the numerical approximations at the same time. In this paper, we will construct third-order conservative sign-preserving Rugne-Kutta and multistep methods to overcome all these difficulties. The time integrations do not depend on the Strang splitting, i.e. we do not split the flux and the stiff source terms. Moreover, the time discretization can handle the stiff source with large time step and preserves the steady-state. Numerical experiments will be given to demonstrate the good performance of the bound-preserving technique and the stability of the scheme for problems with stiff source terms. [ABSTRACT FROM AUTHOR]

This paper presents a topology optimization method in rarefied gas flow problems to obtain the optimal structure of a flow channel as a configuration of gas and solid domains. In this paper, the kinetic equation, the governing equation of rarefied gas flows, is extended over the entire design domain including solid domains assuming the solid as an imaginary gas for implicitly handling the gas-solid interfaces in the optimization process. Based on the extended equation, a 2D flow channel design problem is formulated, and the design sensitivity is obtained based on the Lagrange multiplier method and adjoint variable method. Both the rarefied gas flow and the adjoint flow are computed by a deterministic method based on a finite discretization of the molecular velocity space, rather than the DSMC method. The validity and effectiveness of our proposed method are confirmed through several numerical examples. • We constructed a topology optimization method for rarefied gas flow problems. • The BGK equation was extended over the domain composed of gas and solid. • The sensitivity analysis was performed based on the adjoint variable method. • Numerical examples showed the validity and usefulness of our proposed method. [ABSTRACT FROM AUTHOR]

Computational electrodynamics (CED), the numerical solution of Maxwell's equations, plays an incredibly important role in several problems in science and engineering. High accuracy solutions are desired, and the discontinuous Galerkin (DG) method is one of the better ways of delivering high accuracy in numerical CED. Maxwell's equations have a pair of involution constraints for which mimetic schemes that globally satisfy the constraints at a discrete level are highly desirable. Balsara and Käppeli (2019) presented a von Neumann stability analysis of globally constraint-preserving DG schemes for CED up to fourth order. That paper was focused on developing the theory and documenting the superior dissipation and dispersion of DGTD schemes in media with constant permittivity and permeability. In this paper we present working DGTD schemes for CED that go up to fifth order of accuracy and analyze their performance when permittivity and permeability vary strongly in space. Our DGTD schemes achieve constraint preservation by collocating the electric displacement and magnetic induction as well as their higher order modes in the faces of the mesh. Our first finding is that at fourth and higher orders of accuracy, one has to evolve some zone-centered modes in addition to the face-centered modes. It is well-known that the limiting step in DG schemes causes a reduction of the optimal accuracy of the scheme; though the schemes still retain their formal order of accuracy with WENO-type limiters. In this paper, we document simulations where permittivity and permeability vary by almost an order of magnitude without requiring any limiting of the DG scheme. This very favorable second finding ensures that DGTD schemes retain optimal accuracy even in the presence of large spatial variations in permittivity and permeability. We also study the conservation of electromagnetic energy in these problems. Our third finding shows that the electromagnetic energy is conserved very well even when permittivity and permeability vary strongly in space; as long as the conductivity is zero. • Globally constraint-preserving FR/DG schemes have been designed for CED up to fifth order. • At fourth and higher orders, one has to evolve a few zone-centered modes in addition to the facial modes. • Limiters are not needed even when material properties vary by an order of magnitude. • The high order schemes show excellent conservation of electromagnetic energy even on coarse meshes. [ABSTRACT FROM AUTHOR]

In this paper we propose a novel set of first-order hyperbolic equations that can model dispersive non-hydrostatic free surface flows. The governing PDE system is obtained via a hyperbolic approximation of the family of non-hydrostatic free-surface flow models recently derived by Sainte-Marie et al. in [1]. Our new hyperbolic reformulation is based on an augmented system in which the divergence constraint of the velocity is coupled with the other conservation laws via an evolution equation for the depth-averaged non-hydrostatic pressure, similar to the hyperbolic divergence cleaning applied in generalized Lagrangian multiplier methods (GLM) for magnetohydrodynamics (MHD). We suggest a formulation in which the divergence errors of the velocity field are transported with a large but finite wave speed that is directly related to the maximal eigenvalue of the governing PDE. We then use arbitrary high order accurate (ADER) discontinuous Galerkin (DG) finite element schemes with an a posteriori subcell finite volume limiter in order to solve the proposed PDE system numerically. The final scheme is highly accurate in smooth regions of the flow and very robust and positive preserving for emerging topographies and wet-dry fronts. It is well-balanced making use of a path-conservative formulation of HLL-type Riemann solvers based on the straight line segment path. Furthermore, the proposed ADER-DG scheme with a posteriori subcell finite volume limiter adapts very well to modern GPU architectures, resulting in a very accurate, robust and computationally efficient computational method for non-hydrostatic free surface flows. The new model proposed in this paper has been applied to idealized academic benchmarks such as the propagation of solitary waves, as well as to more challenging physical situations that involve wave runup on a shore including wave breaking in both one and two space dimensions. In all cases the achieved agreement with analytical solutions or experimental data is very good, thus showing the validity of both, the proposed mathematical model and the numerical solution algorithm. • New family of hyperbolic reformulations of depth-averaged models for nonlinear dispersive water waves. • Governing PDE system fulfills an extra energy conservation law (convex extension). • Discretization with arbitrary high order accurate discontinuous Galerkin finite element schemes. • The new approach is computationally efficient and allows large explicit time steps. • Straightforward and highly efficient GPU implementation. [ABSTRACT FROM AUTHOR]

• Deep learning accelerates vapor liquid equilibrium prediction with verified accuracy. • Comprehensive comparison between deep learning and various flash calculation methods. • Optimized deep learning model for estimating VLE properties. • Understandable introduction of deep learning and flash calculation for general audience. • Remarks on further applications of deep learning in classical engineering problems. In the past two decades, researchers have made remarkable progress in accelerating flash calculation, which is very useful in a variety of engineering processes. In this paper, general phase splitting problem statements and flash calculation procedures using the Successive Substitution Method are reviewed, while the main shortages are pointed out. Two acceleration methods, Newton's method and the Sparse Grids Method are presented afterwards as a comparison with the deep learning model proposed in this paper. A detailed introduction from artificial neural networks to deep learning methods is provided here with the authors' own remarks. Factors in the deep learning model are investigated to show their effect on the final result. A selected model based on that has been used in a flash calculation predictor with comparison with other methods mentioned above. It is shown that results from the optimized deep learning model meet the experimental data well with the shortest CPU time. More comparison with experimental data has been conducted to show the robustness of our model. [ABSTRACT FROM AUTHOR]

In this paper, we consider Maxwell's equations in linear dispersive media described by a single-pole Lorentz model for electronic polarization. We study two classes of commonly used spatial discretizations: finite difference methods (FD) with arbitrary even order accuracy in space and high spatial order discontinuous Galerkin (DG) finite element methods. Both types of spatial discretizations are coupled with second order semi-implicit leap-frog and implicit trapezoidal temporal schemes. By performing detailed dispersion analysis for the semi-discrete and fully discrete schemes, we obtain rigorous quantification of the dispersion error for Lorentz dispersive dielectrics. In particular, comparisons of dispersion error can be made taking into account the model parameters, and mesh sizes in the design of the two types of schemes. This work is a continuation of our previous research on energy-stable numerical schemes for nonlinear dispersive optical media [6,7]. The results for the numerical dispersion analysis of the reduced linear model, considered in the present paper, can guide us in the optimal choice of discretization parameters for the more complicated and nonlinear models. The numerical dispersion analysis of the fully discrete FD and DG schemes, for the dispersive Maxwell model considered in this paper, clearly indicate the dependence of the numerical dispersion errors on spatial and temporal discretizations, their order of accuracy, mesh discretization parameters and model parameters. The results obtained here cannot be arrived at by considering discretizations of Maxwell's equations in free space. In particular, our results contrast the advantages and disadvantages of using high order FD or DG schemes and leap-frog or trapezoidal time integrators over different frequency ranges using a variety of measures of numerical dispersion errors. Finally, we highlight the limitations of the second order accurate temporal discretizations considered. • Maxwell's equations. • Lorentz model. • Numerical dispersion. • Finite differences. • Discontinuous Galerkin finite elements. [ABSTRACT FROM AUTHOR]

In this paper we propose a fluctuation splitting finite volume scheme for a non-conservative modeling of shear shallow water flow (SSWF). This model was originally proposed by Teshukov (2007) in [14] and was extended to include modeling of friction by Gavrilyuk et al. (2018) in [7]. The directional splitting scheme proposed by Gavrilyuk et al. (2018) in [7] is tricky to apply on unstructured grids. Our scheme is based on the physical splitting in which we separate the characteristic waves of the model to form two different hyperbolic sub-systems. The fluctuations associated with each sub-systems are computed by developing Riemann solvers for these sub-systems in a local coordinate system. These fluctuations enable us to develop a Godunov-type scheme that can be easily applied on mixed/unstructured grids. While the equation of energy conservation is solved along with the SSWF model in [7] , in this paper we solve only SSWF model equations. We develop a cell-centered finite volume code to validate the proposed scheme with the help of some numerical tests. As expected, the scheme shows first order convergence. The numerical simulation of 1D roll waves shows a good agreement with the experimental results. The numerical simulations of 2D roll waves show similar transverse wave structures as observed in [7]. • A finite volume scheme is proposed for non-conservative shear shallow water equations. • A preferable 'family of paths' is used to design approximate Riemann solvers. • These Riemann solvers are used in to design a fluctuation based Godunov-type scheme. • The proposed scheme is validated by performing some numerical tests. • The scheme can be a useful tool to study the shear shallow water equations. [ABSTRACT FROM AUTHOR]

Nowadays numerical methods for handling partial differential equations is so vast that it is almost taken for granted. In this paper, we propose a discrete duality finite volume scheme with biquadratic elements and its application in solving two-dimensional elliptic equation on quadrilateral meshes. To test the robustness and efficiency of this method, we investigate several examples from physical and engineering fields with different parameters and coefficients. Numerical experiments, encompassing linear and nonlinear elliptic equations with constant or variable coefficients, reveal that the new method achieves optimal convergence in the continuous H 1 -norm, with a 2nd-order convergence, and in the L 2 -norm, with a 3rd-order convergence. Moreover, comparison with the existing biquadratic element finite volume scheme shows that the above method has an advantage of accuracy in the discrete L ∞ -norm on several distorted meshes. [ABSTRACT FROM AUTHOR]

• A second order intersection-based remapping method for cell-centered quantities in two dimensional cylindrical coordinates is proposed. • The remapping method is capable to preserve spherical symmetry when the mesh and mesh moving are symmetric. • The remapping method has good properties such as conservation for mass, momentum and total energy. • The symmetry preserving property and second order accuracy are strictly proved and demonstrated by numerical tests. The importance of preserving physical symmetry in arbitrary Lagrangian-Eulerian (ALE) simulations is well recognized. In the past decades, first order spherical-symmetry-preserving Lagrangian and ALE methods in two-dimensional cylindrical coordinates were developed. And there are also several works about second order spherical-symmetry-preserving cell-centered Lagrangian schemes. The goal of this paper is to develop Second order spherical-symmetry-preserving intersection-based remapping method in two-dimensional cylindrical coordinates. The method is capable to preserve one-dimensional spherical symmetry when the old grid is equiangular polar mesh and the rezoning algorithm only moves points in the radial directions by the same amount for all points with the same spherical radius. The remapping method has second order accuracy almost uniformly (the accuracy drops to first order close to the revolution axis), has good properties such as conservation for mass, momentum and total energy. The symmetry-preserving property and second order accuracy of our new remapping method are proved mathematically and demonstrated by several numerical examples. [ABSTRACT FROM AUTHOR]

In this paper, we develop a class of high-order conservative methods for simulating non-equilibrium radiation diffusion problems. Numerically, this system poses significant challenges due to strong nonlinearity within the stiff source terms and the degeneracy of nonlinear diffusion terms. Explicit methods require impractically small time steps, while implicit methods, which offer stability, come with the challenge to guarantee the convergence of nonlinear iterative solvers. To overcome these challenges, we propose a predictor-corrector approach and design proper implicit-explicit time discretizations. In the predictor step, the system is reformulated into a nonconservative form and linear diffusion terms are introduced as a penalization to mitigate strong nonlinearities. We then employ a Picard iteration to secure convergence in handling the nonlinear aspects. The corrector step guarantees the conservation of total energy, which is vital for accurately simulating the speeds of propagating sharp fronts in this system. For spatial approximations, we utilize local discontinuous Galerkin finite element methods, coupled with positive-preserving and TVB limiters. We validate the orders of accuracy, conservation properties, and suitability of using large time steps for our proposed methods, through numerical experiments conducted on one- and two-dimensional spatial problems. In both homogeneous and heterogeneous non-equilibrium radiation diffusion problems, we attain a time stability condition comparable to that of a fully implicit time discretization. Such an approach is also applicable to many other reaction-diffusion systems. • High order conservative LDG-IMEX methods for non-equilibrium radiation diffusion problems are developed. • Degeneracy, strong nonlinearity and stiff source are all taken into account. • A predictor-corrector procedure based on an IMEX scheme is proposed. • Only mildly nonlinear system needs to be solved and is robust for convergence. • The scheme is of high order, conservative, positivity-preserving and allow large time steps. [ABSTRACT FROM AUTHOR]

The recently-developed general synthetic iterative scheme (GSIS) is efficient in simulating multiscale rarefied gas flows due to the coupling of mesoscopic kinetic equation and macroscopic synthetic equation: for linearized Poiseuille flow where the boundary flux is fixed at each iterative step, the steady-state solutions are found within dozens of iterations in solving the gas kinetic equations, while for general nonlinear flows the iteration number is increased by about one order of magnitude, caused by the incompatible treatment of the boundary flux for the macroscopic synthetic equation. In this paper, we propose a generalized boundary treatment (GBT) to further accelerate the convergence of GSIS. The main idea is, the truncated velocity distribution function at the boundary, similar to that used in the Grad 13-moment equation, is reconstructed by the macroscopic conserved quantities from the synthetic equation, as well as the high-order correction of non-equilibrium stress and heat flux from the kinetic equation; therefore, in each inner iteration solving the synthetic equation, the explicit constitutive relations facilitate real-time updates of the macroscopic boundary flux, driving faster information exchange in the flow field, and consequently achieving faster convergence. Moreover, the high-order correction derived from the kinetic equation can compensate the approximation by the truncation and ensure the boundary accuracy. The one-dimensional Fourier flow, two-dimensional hypersonic flow around a cylinder, three-dimensional pressure-driven pipe flow and the flow around the hypersonic technology vehicle are simulated. The accuracy of GSIS-GBT is validated by the direct simulation Monte Carlo method, the previous versions of GSIS, and the unified gas-kinetic wave-particle method. For the efficiency, in the near-continuum flow regime and slip regime, GSIS-GBT can be faster than the conventional iteration scheme in the discrete velocity method and the previous versions of GSIS by two- and one-order of magnitude, respectively. • The generalized boundary treatment (GBT) is proposed to further accelerate the general synthetic iterative scheme (GSIS). • In both low-speed internal flows and hypersonic flows, GSIS-GBT can achieve convergence within 100 iterations. • GSIS-GBT is respectively faster than previous version of GSIS/conventional iterative scheme by one/two orders of magnitude. [ABSTRACT FROM AUTHOR]

It is increasingly realized that taking stochastic effects into account is important in order to study biological cells. However, the corresponding mathematical formulation, the chemical master equation (CME), suffers from the curse of dimensionality and thus solving it directly is not feasible for most realistic problems. In this paper we propose a dynamical low-rank algorithm for the CME that reduces the dimensionality of the problem by dividing the reaction network into partitions. Only reactions that cross partitions are subject to an approximation error (everything else is computed exactly). This approach, compared to the commonly used stochastic simulation algorithm (SSA, a Monte Carlo method), has the advantage that it is completely noise-free. This is particularly important if one is interested in resolving the tails of the probability distribution. We show that in some cases (e.g. for the lambda phage) the proposed method can drastically reduce memory consumption and run time and provide better accuracy than SSA. • We propose a low-rank algorithm that does not rely on single orbital basis functions. • The proposed approach is better suited for biological/chemical reaction networks. • We demonstrate better accuracy and improved efficiency compared to SSA for some problems. • The proposed complexity reduction algorithm is noise free (in contrast to SSA). • We demonstrate that our algorithm accurately resolves the tails of the distribution (in contrast to SSA). [ABSTRACT FROM AUTHOR]

This paper proposes a new Helmholtz decomposition based windowed Green function (HD-WGF) method for solving the time-harmonic elastic scattering problems on a half-space with Dirichlet boundary conditions in both 2D and 3D. The Helmholtz decomposition is applied to separate the pressure and shear waves, which satisfy the Helmholtz and Helmholtz/Maxwell equations, respectively, and the corresponding boundary integral equations of type (I + T) ϕ = f , that couple these two waves on the unbounded surface, are derived based on the free-space fundamental solution of Helmholtz equation. This approach avoids the treatment of the complex elastic displacement tensor and traction operator that involved in the classical integral equation method for elastic problems. Then a smooth "slow-rise" windowing function is introduced to truncate the boundary integral equations and a "correction" strategy is proposed to ensure the uniformly fast convergence for all incident angles of plane incidence. Numerical experiments for both two and three dimensional problems are presented to demonstrate the accuracy and efficiency of the proposed method. • A novel Helmholtz decomposition based windowed Green function method is proposed for solving the elastic half-space problem. • New boundary integral equations based on Helmholtz decomposition are derived. • The new approach avoids the treatment of the full elastic fundamental tensor and traction operator. • Numerical examples demonstrate the uniform fast convergence for all incident angles of plane incidence. [ABSTRACT FROM AUTHOR]

In this paper, we develop the first entirely graphics processing unit (GPU) based h -adaptive flux reconstruction (FR) method with linear trees. The adaptive solver fully operates on the GPU hardware, using a linear quadtree for two dimensional (2D) problems and a linear octree for three dimensional (3D) problems. We articulate how to efficiently perform tree construction, 2:1 balancing, connectivity query, and how to perform adaptation for the flux reconstruction method on the GPU hardware. As a proof of concept, we apply the adaptive flux reconstruction method to solve the inviscid isentropic vortex propagation problem on 2D and 3D meshes and transitional flow over a 3D square cylinder at Re = 400 to demonstrate the efficiency of the developed adaptive FR method on a single GPU card. Depending on the computational domain size, acceleration of one or two orders of magnitude can be achieved compared to uniform meshing for long distance transportation. The total computational cost of adaption, including tree manipulations, connectivity query and data transfer, compared to that of the numerical solver, is insignificant. It can be less than 2% of the total wall clock time for 3D problems when we perform adaptation every 10-40 time steps with an explicit 3-stage Runge–Kutta time integrator. • Entirely GPU based h -adaptation via linear trees. • One or two orders of magnitude speedup can be achieved. • Adaptation takes as little as two to three percent of the total runtime for 3D simulations. [ABSTRACT FROM AUTHOR]

This paper presents a model for simulating a two-phase flow involving cavitation and condensation, focusing on its application to turbopump flows. The model is derived by extending Kapila's approach, which incorporates thermal and free energy relaxation terms, to compressible multiphase flow equations in a rotating reference frame. The numerical method employed is a finite volume approach based on the Godunov scheme, ensuring pressure equilibrium and total energy conservation through a relaxation step. The model utilizes separate Equations of State (EOS) for the liquid and vapor phases, with the experimental saturation curve, expressed as p sat (T) , being the only parameter needed for cavitation modeling. To broaden its applicability beyond axial pumps, a specific Riemann solver is developed to couple rotating and non-rotating regions within the flow domain. The thermodynamic phase change process is validated using simple cases, and the model is further utilized to simulate water flows in a turbopump inducer. The characteristic curve of the pump is obtained in the non-cavitating regime for several mass flow rates conditions, and the model is successfully assessed in severe cavitation conditions, demonstrating the performance breakdown caused by vapor pockets. The modeling also provides detailed flow descriptions, including analysis of vapor pockets and their impact on blade pressure loading. • Modeling multiphase flow in rotating reference frame. • A new Riemann problem to connect fixed and rotating flow regions. • Phase change modeling by a relaxation method from pure phase to mixture. • Cavitating flow in turbopumps without parameter for the growth of the cavitation pockets. [ABSTRACT FROM AUTHOR]

The present paper focuses on lattice Boltzmann method (LBM) simulations of turbulent flows in combination with a Zonal Detached Eddy Simulation (ZDES) approach, specifically its mode 2 (2020). A common issue of the concomitant use of LBM and DES type models is the systematic modeled-stressed depletion (MSD) triggered by fine isotropic grids imposed by the lattice. This issue is responsible for erroneous surface coefficients that can eventually lead to grid induced separation (GIS). In this work, it is shown that the enhanced boundary layer protection brought by ZDES mode 2 (2020) is sufficient to allow the combination of LBM and a DES type method for high Reynolds number with equal accuracy as its Navier-Stokes counterpart. The emphasis is put on the ability of ZDES to recover its main properties within the numerics of lattice Boltzmann formalism. This is achieved on multiple test-cases namely the turbulent flow over a flat plate, a backward facing step and a three-element airfoil. • Salient combination of lattice Boltzmann Method and Zonal Detached Eddy Simulation. • RANS shielding of turbulent boundary layers ensured even on fine LBM-induced grids. • Consistency between the model and its Navier-Stokes counterpart. • Model assessment on flat plate, backward-facing step and three-element airfoil test-cases. • Robust combination promising for industrial applications. [ABSTRACT FROM AUTHOR]

In this paper, we extend the Generalized Finite Difference Method (GFDM) on unknown compact submanifolds of the Euclidean domain, identified by randomly sampled data that (almost surely) lie on the interior of the manifolds. Theoretically, we formalize GFDM by exploiting a representation of smooth functions on the manifolds with Taylor's expansions of polynomials defined on the tangent bundles. We illustrate the approach by approximating the Laplace-Beltrami operator, where a stable approximation is achieved by a combination of Generalized Moving Least-Squares algorithm and novel linear programming that relaxes the diagonal-dominant constraint for the estimator to allow for a feasible solution even when higher-order polynomials are employed. We establish the theoretical convergence of GFDM in solving Poisson PDEs and numerically demonstrate the accuracy on simple smooth manifolds of low and moderate high co-dimensions as well as unknown 2D surfaces. For the Dirichlet Poisson problem where no data points on the boundaries are available, we employ GFDM with the volume-constraint approach that imposes the boundary conditions on data points close to the boundary. When the location of the boundary is unknown, we introduce a novel technique to detect points close to the boundary without needing to estimate the distance of the sampled data points to the boundary. We demonstrate the effectiveness of the volume-constraint employed by imposing the boundary conditions on the data points detected by this new technique compared to imposing the boundary conditions on all points within a certain distance from the boundary, where the latter is sensitive to the choice of truncation distance and requires the knowledge of the boundary location. [ABSTRACT FROM AUTHOR]

This paper proposes a deep-learning-based method for recovering a signed distance function (SDF) of a given hypersurface represented by an implicit level set function. Using the flexibility of constructing a neural network, we use an augmented network by defining an auxiliary output to represent the gradient of the SDF. There are three advantages of the augmented network; (i) the target interface is accurately captured, (ii) the gradient has a unit norm, and (iii) two outputs are approximated by a single network. Moreover, unlike a conventional loss term which uses a residual of the eikonal equation, a novel training objective consisting of three loss terms is designed. The first loss function enforces a pointwise matching between two outputs of the augmented network. The second loss function leveraged by a geometric characteristic of the SDF imposes the shortest path obtained by the gradient. The third loss function regularizes a singularity of the SDF caused by discontinuities of the gradient. Numerical results across a wide range of complex and irregular interfaces in two and three-dimensional domains confirm the effectiveness and accuracy of the proposed method. We also compare the results of the proposed method with physics-informed neural networks approaches and the fast marching method. • A deep-learning-based method is proposed to recover a signed distance function (SDF) of a level set function. • An augmented neural network structure with the gradient of the SDF as an auxiliary output is designed. • From geometric properties, new loss functions are proposed that impose global relations and alleviate the singularity of SDF. • Several numerical tests on irregular interfaces in 2D and 3D validate the effectiveness and accuracy of the proposed method. [ABSTRACT FROM AUTHOR]

This paper investigates an efficient exponential integrator generalized multiscale finite element method for solving a class of time-evolving partial differential equations in bounded domains. The proposed method first performs the spatial discretization of the model problem using constraint energy minimizing generalized multiscale finite element method (CEM-GMsFEM). This approach consists of two stages. First, the auxiliary space is constructed by solving local spectral problems, where the basis functions corresponding to small eigenvalues are captured. The multiscale basis functions are obtained in the second stage using the auxiliary space by solving local energy minimization problems over the oversampling domains. The basis functions have exponential decay outside the corresponding local oversampling regions. We shall consider the first and second-order explicit exponential Runge-Kutta approach for temporal discretization and to build a fully discrete numerical solution. The exponential integration strategy for the time variable allows us to take full advantage of the CEM-GMsFEM as it enables larger time steps due to its stability properties. We derive the error estimates in the energy norm under the regularity assumption. Finally, we will provide some numerical experiments to sustain the efficiency of the proposed method. • Multiscale reduction methods are efficient and accurate to solve problems with high-contrast media. • The presence of high-contrast coefficients reduces the stability region of time discretization. • Implicit methods must solve nonlinear equations, which can be a bottleneck computation. • Exponential integrators are alternative techniques and use robust time-stepping. [ABSTRACT FROM AUTHOR]

The simulation of plasma physics is computationally expensive because the underlying physical system is of high dimensions, requiring three spatial dimensions and three velocity dimensions. One popular numerical approach is Particle-In-Cell (PIC) methods owing to its ease of implementation and favorable scalability in high-dimensional problems. An unfortunate drawback of the method is the introduction of statistical noise resulting from the use of finitely many particles. In this paper we examine the application of the Smoothness-Increasing Accuracy-Conserving (SIAC) family of convolution kernel filters as denoisers for moment data arising from PIC simulations. We show that SIAC filtering is a promising tool to denoise PIC data in the physical space as well as capture the appropriate scales in the Fourier space. Furthermore, we demonstrate how the application of the SIAC technique reduces the amount of information necessary in the computation of quantities of interest in plasma physics such as the Bohm speed. • SIAC filters are effective denoisers for moment data arising from PIC simulations. • SIAC captures appropriate Fourier signal information. • SIAC reduces amount of information necessary in computation of quantities of interest. • SIAC is useful in Bohm speed calculation. [ABSTRACT FROM AUTHOR]

Steepest descent methods combining complex contour deformation with numerical quadrature provide an efficient and accurate approach for the evaluation of highly oscillatory integrals. However, unless the phase function governing the oscillation is particularly simple, their application requires a significant amount of a priori analysis and expert user input, to determine the appropriate contour deformation, and to deal with the non-uniformity in the accuracy of standard quadrature techniques associated with the coalescence of stationary points (saddle points) with each other, or with the endpoints of the original integration contour. In this paper we present a novel algorithm for the numerical evaluation of oscillatory integrals with general polynomial phase functions, which automates the contour deformation process and avoids the difficulties typically encountered with coalescing stationary points and endpoints. The inputs to the algorithm are simply the phase and amplitude functions, the endpoints and orientation of the original integration contour, and a small number of numerical parameters. By a series of numerical experiments we demonstrate that the algorithm is accurate and efficient over a large range of frequencies, even for examples with a large number of coalescing stationary points and with endpoints at infinity. As a particular application, we use our algorithm to evaluate cuspoid canonical integrals from scattering theory. A Matlab implementation of the algorithm is made available and is called PathFinder. • An algorithm for the evaluation of oscillatory integrals with polynomial phase. • The algorithm automates the numerical steepest descent method. • Our implementation is robust and efficient across all frequencies. [ABSTRACT FROM AUTHOR]

In the paper we construct one extended Filon method for the multivariate oscillatory integral on a regular simplex avoiding the derivatives. Based on the proposed geometric configuration, the error estimate of the multivariate Lagrange interpolation on a simplex is established. Based on the multivariate interpolating, the extended Filon method is presented to reduce the integration error for the whole regime of the oscillatory parameter. In particular, the explicit formula of the moment is also provided. Numerical results illustrate that the error decreases with increasing oscillatory parameters and more interpolating nodes. [ABSTRACT FROM AUTHOR]

In this paper, we propose and numerically solve a model considering confined flow in fractured porous media coupled with free flow in conduits. Such a situation arises, for example, for fluid flows in fractured vuggy carbonate reservoirs and caves/conduits, where a matrix and fractures are the main spaces for fluid storage and flow. The flow in caves and conduits is governed by the Stokes equation; the flow in fractured porous media, which consists of both a matrix and fractures, is described by a coupled Darcy–Darcy (bulk–fracture) model. Then, the two models are coupled through suitable (physically consistent) interface conditions on the interface between fractured porous media and caves/conduits, playing a key role in a physically faithful simulation that achieves high accuracy. The construction of all the interface conditions of the coupled model is based on the fundamental properties of the traditional bulk–fracture model and the well-known Stokes–Darcy model. The weak formulation is derived for the proposed model, and the well-posedness of the model is theoretically analyzed. We also propose a polyhedral discontinuous Galerkin finite element approximation for the Stokes flow in conduits and Darcy flow in the porous matrix, which is coupled with a conforming finite element scheme for the flow in the fracture. Optimal error estimates in a suitable energy norm are obtained. Two-dimensional numerical experiments are conducted to validate the proposed model and to demonstrate the features of both the model and the numerical method, such as the optimal convergence rate of the numerical solution; the detail flow characteristics around fractures, the matrix, and conduits; and the applicability to real-world problems. [ABSTRACT FROM AUTHOR]

This paper investigates an inverse scattering problem of time-harmonic elastic waves incident on a rigid random periodic structure. A novel and efficient numerical method is proposed to quantify the randomness, i.e., to reconstruct key statistical properties of random structures from boundary measurements of the scattering data. This method consists of two ingredients, the Monte Carlo technique for sampling the probability space and a continuation method with respect to the wavenumber. Unlike existing methods, this approach does not require a priori knowledge of the randomness and can handle a wide range of random structures, including non-Gaussian processes with extremely high levels of complexity. To illustrate the accuracy and effectiveness of the proposed method, numerical examples involving Gaussian and non-Gaussian processes with various covariances are provided. [ABSTRACT FROM AUTHOR]

Accurate and efficient numerical simulations of soil water movement, as described by the highly nonlinear Richards' equation, often require local refinement near recharge or sink/source terms. In this paper, we present a novel numerical scheme for solving Richards' equation. Our approach is based on the vertex-centered finite volume method (VCFVM) and can be easily adapted to locally refined meshes. The proposed scheme offers some key features, including the definition of all unknowns over vertices of the primary mesh, expression of flux crossing dual edges as combinations of hydraulic heads at the vertices of the primary cell, and the capability to handle nonmatching meshes in the presence of local mesh refinement. For performance evaluation, soil water content and soil water potential simulated by the proposed scheme are benchmarked against results produced from HYDRUS (a widely used soil water numerical model) and the observed values in four test cases, including a convergence test case, a synthetic case, a laboratory experiment case and a field experiment case. The comparison results demonstrate the effectiveness and applicability of our scheme across a wide range of soil parameters and boundary conditions. [ABSTRACT FROM AUTHOR]

The algebraic volume-of-fluid (VOF) method based on the construction strategy of switching technique has received extensive attention recently. However, all the previous algebraic VOF methods were constructed by combining different methods of compressible differencing scheme (CDS) and the high resolution scheme (HRS), which would switch back and forth in most cases. Thus the final algebraic VOF method usually becomes to be complicated, and the numerical instabilities would be increased. In this paper, a novel algebraic VOF method, without the operations of switching back and forth, is constructed within a unified framework of the steepness-adjustable harmonic (SAH) scheme (He et al. (2022) [33]). A thorough validation of the present method is conducted, examining the pure advection of the interface indicator function. The results indicate that the present method can resolve the interface capturing with substantially low numerical diffusion and low numerical oscillations. • The same expression of SAH scheme is utilized for high resolution and compressible differencing scheme with only different steepness parameter. • The switching technique in the classical algebraic VOF method is extended to get the final steepness parameter based on the infimum and supremum. • A strategy based on least squares approximation parameter is proposed for determining the supremum of the steepness. [ABSTRACT FROM AUTHOR]

• We proposed variable linear transformation improved physics-informed neural networks (VLT-PINNs) to solve thin-layer flow forward and inverse problems. • Three principles for determining the VLT parameters were identified for three types of variables in PINNs respectively. • The VLT-PINNs with the principle-suggested parameters show excellent performance over the ten test cases. • Only under the constraints of the VLT principles can we obtain satisfactory results, while traditional linear transformation may yield totally wrong results. • VLT method is an attempt to combine the three advantages of accuracy, universality and simplicity. Physics-informed neural networks (PINNs) have attracted wide attention due to their ability to seamlessly embed the learning process with physical laws and their considerable success in solving forward and inverse differential equation (DE) problems. While most studies are improving the learning process and network architecture of PINNs, less attention has been paid to the modification of the DE system, which may play an important role in addressing some limitations of PINNs. One of the simplest modifications that can be implemented to all DE systems is the variable linear transformation (VLT). Therefore, in this work, we propose the VLT-PINNs that solve the DE systems of the linear-transformed variables instead of the original ones. To clearly illustrate the importance of prior knowledge in determining the VLT parameters, we choose the thin-layer flow problems as our focus. Ten related cases were tested, including the jet flows, wake flows, mixing layers, boundary layers and Kovasznay flows. Based on the principle of normalization and for a better match of the DE system to the preference of NNs, we identify three principles for determining the VLT parameters: magnitude normalization for dependent variables (principle 1), local normalization for independent variables (principle 2), and appropriate scaling for physics-related parameters in inverse problems (principle 3). The VLT-PINNs with the VLT parameters suggested by the proposed principles show excellent performance over all the test cases, while the results are quite poor with the VLT parameters suggested by traditional linear transformations, such as nondimensionalization and global normalization. Comparison studies also show that only under the constraints of the VLT principles can we obtain satisfactory results. Besides, we find tanh is more appropriate as the activation function than sin for thin-layer flow problems, from both posteriori results and priori analyses with physical intuition. We highlight that our VLT method is an attempt to combine the three advantages of accuracy, universality and simplicity, and hope that it can provide new insights into the better integration of prior knowledge, physical intuition and the nature of NNs. The code for this paper is available on https://github.com/CAME-THU/VLT-PINN. [ABSTRACT FROM AUTHOR]

We propose an easy-to-implement iterative method for resolving the implicit (or semi-implicit) schemes arising in solving reaction-diffusion (RD) type equations. We formulate the nonlinear time implicit scheme as a min-max saddle point problem and then apply the primal-dual hybrid gradient (PDHG) method. Suitable precondition matrices are applied to the PDHG method to accelerate the convergence of algorithms under different circumstances. Furthermore, our method is applicable to various discrete numerical schemes with high flexibility. From various numerical examples tested in this paper, the proposed method converges properly and can efficiently produce numerical solutions with sufficient accuracy. [ABSTRACT FROM AUTHOR]

Fourier analysis based optimization techniques have been developed for numerical schemes on structured grids and result in significant performance improvements. However, there lacks an effective optimization technique for numerical schemes on general unstructured grids, due to the mesh geometry complexity that makes Fourier transformation infeasible. To address this issue, an optimization framework based on machine learning is developed in this paper for numerical schemes on unstructured grids. The optimization of a compact high-order variational reconstruction on triangular grids is performed to illustrate the framework. An artificial neural network (ANN) is used to predict the optimal values of the derivative weights on cell interfaces that are the free parameters of the variational reconstruction, given the local geometry as input. The ANN is trained by minimizing the solution errors of a linear advection equation on a set of generated compact stencils, with a set of sampled sine waves as initial conditions. The developed optimization framework is applicable to general numerical schemes on unstructured grids as the training does not require explicit computation of dispersion or dissipation. Numerical results for inviscid flow problems show that the optimized variational finite volume scheme is remarkably more accurate and efficient than its non-optimized counterpart, demonstrating the effectiveness of the machine learning optimization. • A machine learning optimization framework for numerical schemes on unstructured grids. • A feedforward neural network to predict optimal parameter values of numerical schemes. • Training of the network by minimizing solution errors of a linear advection equation. • A machine learning optimized compact high-order variational reconstruction. • Generalization of the optimized variational reconstruction to nonlinear problems. [ABSTRACT FROM AUTHOR]

Stochastic Klein–Gordon–Schrödinger (KGS) equations are important mathematical models and describe the interaction between scalar nucleons and neutral scalar mesons in the stochastic environment. In this paper, we propose novel structure-preserving schemes to numerically solve stochastic KGS equations with additive noise, which preserve averaged charge evolution law, averaged energy evolution law, symplecticity, and multi-symplecticity. By applying central difference, sine pseudo-spectral method, or finite element method in space and modifying finite difference in time, we present some charge and energy preserved fully-discrete schemes for the original system. In addition, combining the symplectic Runge-Kutta method in time and finite difference in space, we propose a class of multi-symplectic discretizations preserving the geometric structure of the stochastic KGS equation. Finally, numerical experiments confirm theoretical findings. [ABSTRACT FROM AUTHOR]

Pseudo-Hamiltonian neural networks (PHNN) were recently introduced for learning dynamical systems that can be modelled by ordinary differential equations. In this paper, we extend the method to partial differential equations. The resulting model is comprised of up to three neural networks, modelling terms representing conservation, dissipation and external forces, and discrete convolution operators that can either be learned or be given as input. We demonstrate numerically the superior performance of PHNN compared to a baseline model that models the full dynamics by a single neural network. Moreover, since the PHNN model consists of three parts with different physical interpretations, these can be studied separately to gain insight into the system, and the learned model is applicable also if external forces are removed or changed. • We introduce pseudo-Hamiltonian neural networks for partial differential equations. • The models can be separated into different parts, with physical interpretations. • Models learned on a disturbed system can predict future states without disturbances. • Assumptions on geometric structures can be imposed on the models. • Our new models consistently outperform the baseline neural network model. [ABSTRACT FROM AUTHOR]