Colloquiam: Documents published in 2020

Phase-field modeling of fracture in linear thin shells

María Jesús Samper — Wed, 26 Feb 2020 14:01:14 +0100

We present a phase-field model for fracture in Kirchoff-Love thin shells using the local maximum-entropy (LME) meshfree method. Since the crack is a natural outcome of the analysis it does not require an explicit representation and tracking, which is advantage over techniques as the extended finite element method that requires tracking of the crack paths. The geometric description of the shell is based on statistical learning techniques that allow dealing with general point set surfaces avoiding a global parametrization, which can be applied to tackle surfaces of complex geometry and topology. We show the flexibility and robustness of the present methodology for two examples: plate in tension and a set of open connected pipes.

XLME interpolants, a seamless bridge between XFEM and enriched meshless methods

María Jesús Samper — Wed, 26 Feb 2020 13:53:35 +0100

In this paper, we develop a method based on local maximum entropy shape functions together with enrichment functions used in partition of unity methods to discretize problems in linear elastic fracture mechanics. We obtain improved accuracy relative to the standard extended finite element method at a comparable computational cost. In addition, we keep the advantages of the LME shape functions, such as smoothness and non-negativity. We show numerically that optimal convergence (same as in FEM) for energy norm and stress intensity factors can be obtained through the use of geometric (fixed area) enrichment with no special treatment of the nodes near the crack such as blending or shifting.

Shape control of active surfaces inspired by the movement of euglenids

María Jesús Samper — Wed, 26 Feb 2020 13:35:27 +0100

We examine a novel mechanism for active surface morphing inspired by the cell body deformations of euglenids. Actuation is accomplished through in-plane simple shear along prescribed slip lines decorating the surface. Under general non-uniform actuation, such local deformation produces Gaussian curvature, and therefore leads to shape changes. Geometrically, a deformation that realizes the prescribed local shear is an isometric embedding. We explore the possibilities and limitations of this bio- inspired shape morphing mechanism, by first characterizing isometric embeddings un- der axisymmetry, understanding the limits of embeddability, and studying in detail the accessibility of surfaces of zero and constant curvature. Modeling mechanically the active surface as a non-Euclidean plate (NEP), we further examine the mechanism beyond the geometric singularities arising from embeddability, where mechanics and buckling play a decisive role. We also propose a non-axisymmetric actuation strategy to accomplish large amplitude bending and twisting motions of elongated cylindrical surfaces. Besides helping understand how euglenids delicately control their shape, our results may provide the background to engineer soft machines

Nonsingular isogeometric boundary element method for stokes flows in 3D

María Jesús Samper — Wed, 26 Feb 2020 13:28:09 +0100

Isogeometric analysis (IGA) is emerging as a technology bridging computer aided geometric design (CAGD), most commonly based on Non-Uniform Rational B-Splines (NURBS) surfaces, and engineering analysis. In finite element and boundary element isogeometric methods (FE-IGA and IGA-BEM), the NURBS basis functions that describe the geometry define also the approximation spaces. In the FE-IGA approach, the surfaces generated by the CAGD tools need to be extended to volumetric descriptions, a major open problem in 3D. This additional passage can be avoided in principle when the partial differential equations to be solved admit a formulation in terms of boundary integral equations, leading to boundary element isogeometric analysis (IGA-BEM). The main advantages of such an approach are given by the dimensionality reduction of the problem (from volumetric-based to surface-based), by the fact that the interface with CAGD tools is direct, and by the possibility to treat exterior problems, where the computational domain is infinite. By contrast, these methods produce system matrices which are full, and require the integration of singular kernels. In this paper we address the second point and propose a nonsingular formulation of IGA-BEM for 3D Stokes flows, whose convergence is carefully tested numerically. Standard Gaussian quadrature rules suffice to integrate the boundary integral equations, and carefully chosen known exact solutions of the interior Stokes problem are used to correct the resulting matrices, extending the work by Klaseboer et al. (2012) [27] to IGA-BEM.

An adaptive meshfree method for phase-field models of biomembranes. Part II: A Lagrangian approach for membranes in viscous fluids

María Jesús Samper — Wed, 26 Feb 2020 13:14:33 +0100

We present a Lagrangian phase-field method to study the low Reynolds number dynamics of vesicles embedded in a viscous fluid. In contrast to previous approaches, where the field variables are the phase-field and the fluid velocity, here we exploit the fact that the phasefield tracks a material interface to reformulate the problem in terms of the Lagrangian motion of a background medium, containing both the biomembrane and the fluid. We discretize the equations in space with maximum-entropy approximants, carefully shown to perform well in phase-field models of biomembranes in a companion paper. The proposed formulation is variational, lending itself to implicit time-stepping algorithms based on minimization of a time-incremental energy, which are automatically nonlinearly stable. The proposed method deals with two of the major challenges in the numerical treatment of coupled fluid/phase-field models of biomembranes, namely the adaptivity of the grid to resolve the sharp features of the phase-field, and the stiffness of the equations, leading to very small time-steps. In our method, local refinement follows the features of the phasefield as both are advected by the Lagrangian motion, and large time-steps can be robustly chosen in the variational time-stepping algorithm, which also lends itself to time adaptivity. The method is presented in the axisymmetric setting, but it can be directly extended to 3D. We present a Lagrangian phase-field method to study the low Reynolds number dynamics of vesicles embedded in a viscous fluid. In contrast to previous approaches, where the field variables are the phase-field and the fluid velocity, here we exploit the fact that the phase-field tracks a material interface to reformulate the problem in terms of the Lagrangian motion of a background medium, containing both the biomembrane and the fluid. We discretize the equations in space with maximum-entropy approximants, carefully shown to perform well in phase-field models of biomembranes in a companion paper. The proposed formulation is variational, lending itself to implicit time-stepping algorithms based on minimization of a time-incremental energy, which are automatically nonlinearly stable. The proposed method deals with two of the major challenges in the numerical treatment of coupled fluid/phase-field models of biomembranes, namely the adaptivity of the grid to resolve the sharp features of the phase-field, and the stiffness of the equations, leading to very small time-steps. In our method, local refinement follows the features of the phase-field as both are advected by the Lagrangian motion, and large time-steps can be robustly chosen in the variational time-stepping algorithm, which also lends itself to time adaptivity. The method is presented in the axisymmetric setting, but it can be directly extended to 3D.

Blending isogeometric analysis and local maximum entropy meshfree approximants

María Jesús Samper — Wed, 26 Feb 2020 12:55:21 +0100

We present a method to blend local maximum entropy (LME) meshfree approximants and isogeometric analysis. The coupling strategy exploits the optimization program behind LME approximation, treats isogeometric and LME basis functions on an equal footing in the reproducibility constraints, but views the former as data in the constrained minimization. The resulting scheme exploits the best features and overcomes the main drawbacks of each of these approximants. Indeed, it preserves the high fidelity boundary representation (exact CAD geometry) of isogeometric analysis, out of reach for bare meshfree methods, and easily handles volume discretization and unstructured grids with possibly local refinement, while maintaining the smoothness and non-negativity of the basis functions. We implement the method with B-Splines in two dimensions, but the procedure carries over to higher spatial dimensions or to other non-negative approximants such as NURBS or subdivision schemes. The performance of the method is illustrated with the heat equation, and linear and nonlinear elasticity. The ability of the proposed method to impose directly essential boundary conditions in non-convex domains, and to deal with unstructured grids and local refinement in domains of complex geometry and topology is highlighted by the numerical examples.

An adaptive meshfree method for phase-field models of biomembranes. Part I: Approximation with maximum-entropy basis functions

María Jesús Samper — Wed, 26 Feb 2020 12:07:30 +0100

We present an adaptive meshfree method to approximate phase-field models of biomembranes. In such models, the Helfrich curvature elastic energy, the surface area, and the enclosed volume of a vesicle are written as functionals of a continuous phase-field, which describes the interface in a smeared manner. Such functionals involve up to second-order spatial derivatives of the phase-field, leading to fourth-order Euler–Lagrange partial differential equations (PDE). The solutions develop sharp internal layers in the vicinity of the putative interface, and are nearly constant elsewhere. Thanks to the smoothness of the local maximum-entropy (max-ent) meshfree basis functions, we approximate numerically this high-order phase-field model with a direct Ritz–Galerkin method. The flexibility of the meshfree method allows us to easily adapt the grid to resolve the sharp features of the solutions. Thus, the proposed approach is more efficient than common tensor product methods (e.g. finite differences or spectral methods), and simpler than unstructured Cº finite element methods, applicable by reformulating the model as a system of second-order PDE. The proposed method, implemented here under the assumption of axisymmetry, allows us to show numerical evidence of convergence of the phase-field solutions to the sharp interface limit as the regularization parameter approaches zero. In a companion paper, we present a Lagrangian method based on the approximants analyzed here to study the dynamics of vesicles embedded in a viscous fluid. We present an adaptive meshfree method to approximate phase-field models of biomembranes. In such models, the Helfrich curvature elastic energy, the surface area, and the enclosed volume of a vesicle are written as functionals of a continuous phase-field, which describes the interface in a smeared manner. Such functionals involve up to second-order spatial derivatives of the phase-field, leading to fourth-order Euler–Lagrange partial differential equations (PDE). The solutions develop sharp internal layers in the vicinity of the putative interface, and are nearly constant elsewhere. Thanks to the smoothness of the local maximum-entropy (max-ent) meshfree basis functions, we approximate numerically this high-order phase-field model with a direct Ritz–Galerkin method. The flexibility of the meshfree method allows us to easily adapt the grid to resolve the sharp features of the solutions. Thus, the proposed approach is more efficient than common tensor product methods (e.g. finite differences or spectral methods), and simpler than unstructured C0C0 finite element methods, applicable by reformulating the model as a system of second-order PDE. The proposed method, implemented here under the assumption of axisymmetry, allows us to show numerical evidence of convergence of the phase-field solutions to the sharp interface limit as the regularization parameter approaches zero. In a companion paper, we present a Lagrangian method based on the approximants analyzed here to study the dynamics of vesicles embedded in a viscous fluid.

Curved fluid membranes behave laterally as an effective viscoelastic medium

María Jesús Samper — Wed, 26 Feb 2020 12:01:57 +0100

The lateral mobility of membrane inclusions is essential in biological processes involving membrane-bound macromolecules, which often take place in highly curved geometries such as membrane tubes or small organelles. Probe mobility is assisted by the lateral fluidity, which is thought to be purely viscous for lipid bilayers and synthetic systems such as polymersomes. In previous theoretical studies, the hydrodynamical mobility is estimated assuming a fixed membrane geometry. However, fluid membranes are very flexible out-of-plane. By accounting for the deformability of the membrane and in the presence of curvature, we show that the lateral motion of an inclusion produces a normal force, which results in a nonuniform membrane deformation. Such a deformation mobilizes the bending elasticity, produces extra lateral viscous and elastic forces, and results in an effective lateral viscoelastic behavior. The coupling between lateral and out-of-plane mechanics is mediated by the interfacial hydrodynamics and curvature. We analyze the frequency and curvature dependent rheology of flexible fluid membranes, and interpret it with a simple four-element model, which provides a background for microrheological experiments. Two key technical aspects of the present work are a new formulation for the interfacial hydrodynamics, and the linearization of the governing equations around a cylindrical geometry.

Adhesion and friction control localized folding in supported graphene

María Jesús Samper — Wed, 26 Feb 2020 11:57:34 +0100

Graphene deposited on planar surfaces often exhibits sharp and localized folds delimiting seemingly planar regions, as a result of compressive stresses transmitted by the substrate. Such folds alter the electronic and chemical properties of graphene, and therefore, it is important to understand their emergence, to either suppress them or control their morphology. Here, we study the emergence of out-of-plane deformations in supported and laterally strained graphene with high-fidelity simulations and a simpler theoretical model. We characterize the onset of buckling and the nonlinear behavior after the instability in terms of the adhesion and frictional material parameters of the graphene-substrate interface. We find that localized folds evolve from a distributed wrinkling linear instability due to the nonlinearity in the van der Waals graphene-substrate interactions. We identify friction as a selection mechanism for the separation between folds, as the formation of far apart folds is penalized by the work of friction. Our systematic analysis is a first step towards strain engineering of supported graphene, and is applicable to other compressed thin elastic films weakly coupled to a substrate.

Second order convex maximum entropy approximants with applications to high order PDE

María Jesús Samper — Wed, 26 Feb 2020 11:49:03 +0100

We present a new approach for second order maximum entropy (max-ent) meshfree approximants that produces positive and smooth basis functions of uniform aspect ratio even for non-uniform node sets, and prescribes robustly feasible constraints for the entropy maximization program defining the approximants. We examine the performance of the proposed approximation scheme in the numerical solution by a direct Galerkin method of a number of partial differential equations (PDEs), including structural vibrations, elliptic second order PDEs, and fourth order PDEs for Kirchhoff-Love thin shells and for a phase field model describing the mechanics of biomembranes. The examples highlight the ability of the method to deal with non-uniform node distributions, and the high accuracy of the solutions. Surprisingly, the first order meshfree max-ent approximants with large supports are competitive when compared to the proposed second order approach in all the tested examples, even in the higher order PDEs.

Nonlinear manifold learning for meshfree finite deformation thin shell analysis

María Jesús Samper — Wed, 26 Feb 2020 11:43:36 +0100

Calculations on general point-set surfaces are attractive because of their flexibility and simplicity in the preprocessing but present important challenges. The absence of a mesh makes it nontrivial to decide if two neighboring points in the three-dimensional embedding are nearby or rather far apart on the manifold. Furthermore, the topology of surfaces is generally not that of an open two-dimensional set, ruling out global parametrizations. We propose a general and simple numerical method analogous to the mathematical theory of manifolds, in which the point-set surface is described by a set of overlapping charts forming a complete atlas. We proceed in four steps: (1) partitioning of the node set into subregions of trivial topology; (2) automatic detection of the geometric structure of the surface patches by nonlinear dimensionality reduction methods; (3) parametrization of the surface using smooth meshfree (here maximum-entropy ) approximants; and (4) gluing together the patch representations by means of a partition of unity. Each patch may be viewed as a meshfree macro-element. We exemplify the generality, flexibility, and accuracy of the proposed approach by numerically approximating the geometrically nonlinear Kirchhoff–Love theory of thin-shells. We analyze standard benchmark tests as well as point-set surfaces of complex geometry and topology.

Confined bilayers passively regulate shape and stress

María Jesús Samper — Wed, 26 Feb 2020 11:37:14 +0100

Lipid membranes are commonly confined to adjacent subcellular structures or to artificial substrates and particles. We develop an experimental and theoretical framework to investigate the mechanics of confined membranes, including the influence of adhesion, strain, and osmotic pressure. We find that supported lipid bilayers respond to stress by nucleating and evolving spherical and tubular protrusions. In cells, such transformations are generally attributed to proteins. Our results offer insights into the mechanics of cell membranes and can further extend the applications of supported bilayers.

An atomistic-based foliation model for multilayer graphene materials and nanotubes

María Jesús Samper — Wed, 26 Feb 2020 11:02:13 +0100

We present a three-dimensional continuum model for layered crystalline materials made out of weakly interacting two-dimensional crystalline sheets. We specialize the model to multilayer graphene materials, including multi-walled carbon nanotubes (MWCNTs). We view the material as a foliation, partitioning of space into a continuous stack of leaves, thus loosing track of the location of the individual graphene layers. The constitutive model for the bulk is derived from the atomistic interactions by appropriate kinematic assumptions, adapted to the foliation structure and mechanics. In particular, the elastic energy along the leaves of the foliation results from the bonded interactions, while the interaction energy between the walls, resulting from van der Waals forces, is parametrized with a stretch transversal to the foliation. The resulting theory is distinct from conventional anisotropic models, and can be readily discretized with finite elements. The discretization is not tied to the individual walls and allows us to coarse-grain the system in all directions. Furthermore, the evaluation of the non-bonded interactions becomes local. We test the accuracy of the foliation model against a previously proposed atomistic-based continuum model that explicitly describes each and every wall. We find that the new model is very efficient and accurate. Furthermore, it allows us to rationalize the rippling deformation modes characteristic of thick MWCNTs, highlighting the role of the van der Waals forces and the sliding between the walls. By exercising the model with very large systems of hollow MWCNTs and suspended multilayer graphene, containing up to 109 atoms, we find new complex post-buckling deformation patterns.

Modeling and enhanced sampling of molecular systems with smooth and nonlinear data-driven collective variables

María Jesús Samper — Wed, 26 Feb 2020 10:57:04 +0100

Collective variables (CVs) are low-dimensional representations of the state of a complex system, which help us rationalize molecular conformations and sample free energy landscapes with molecular dynamics simulations. Given their importance, there is need for systematic methods that effectively identify CVs for complex systems. In recent years, nonlinear manifold learning has shown its ability to automatically characterize molecular collective behavior. Unfortunately, these methods fail to provide a differentiable function mapping high-dimensional configurations to their low-dimensional representation, as required in enhanced sampling methods. We introduce a methodology that, starting from an ensemble representative of molecular flexibility, builds smooth and nonlinear data-driven collective variables (SandCV) from the output of nonlinear manifold learning algorithms. We demonstrate the method with a standard benchmark molecule, alanine dipeptide, and show how it can be non-intrusively combined with off-the-shelf enhanced sampling methods, here the adaptive biasing force method. We illustrate how enhanced sampling simulations with SandCV can explore regions that were poorly sampled in the original molecular ensemble. We further explore the transferability of SandCV from a simpler system, alanine dipeptide in vacuum, to a more complex system, alanine dipeptide in explicit water.

Nonlinear manifold learning for model reduction in finite elastodynamics

María Jesús Samper — Wed, 26 Feb 2020 10:51:22 +0100

Model reduction in computational mechanics is generally addressed with linear dimensionality reduction methods such as Principal Components Analysis (PCA). Hypothesizing that in many applications of interest the essential dynamics evolve on a nonlinear manifold, we explore here reduced order modeling based on nonlinear dimen- sionality reduction methods. Such methods are gaining popularity in diverse fields of science and technology, such as machine perception or molecular simulation. We consider finite deformation elastodynamics as a model problem, and identify the manifold where the dynamics essentially take place –the slow manifold– by nonlinear dimensionality reduction methods applied to a database of snapshots. Contrary to linear dimensionality reduction, the smooth parametrization of the slow manifold needs special techniques, and we use local maximum entropy approximants. We then formulate the Lagrangian mechanics on these data-based generalized coordinates, and de- velop variational time-integrators. Our proof-of-concept example shows that a few nonlinear collective variables provide similar accuracy to tens of PCA modes, suggesting that the proposed method may be very attractive in control or optimization applications. Furthermore, the reduced number of variables brings insight into the me- chanics of the system under scrutiny. Our simulations also highlight the need of modeling the net e ¿ ect of the disregarded degrees of freedom on the reduced dynamics at long times.

Reverse engineering the euglenoid movement

María Jesús Samper — Wed, 26 Feb 2020 10:43:31 +0100

Euglenids exhibit an unconventional motility strategy amongst unicellular eukaryotes, consisting of large-amplitude highly concerted deformations of the entire body (euglenoid movement or metaboly). A plastic cell envelope called pellicle mediates these deformations. Unlike ciliary or flagellar motility, the biophysics of this mode is not well understood, including its efficiency and molecular machinery. We quantitatively examine video recordings of four euglenids executing such motions with statistical learning methods. This analysis reveals strokes of high uniformity in shape and pace. We then interpret the observations in the light of a theory for the pellicle kinematics, providing a precise understanding of the link between local actuation by pellicle shear and shape control. We systematically understand common observations, such as the helical conformations of the pellicle, and identify previously unnoticed features of metaboly. While two of our euglenids execute their stroke at constant body volume, the other two exhibit deviations of about 20% from their average volume, challenging current models of low Reynolds number locomotion. We find that the active pellicle shear deformations causing shape changes can reach 340%, and estimate the velocity of the molecular motors. Moreover, we find that metaboly accomplishes locomotion at hydrodynamic efficiencies comparable to those of ciliates and flagellates. Our results suggest new quantitative experiments, provide insight into the evolutionary history of euglenids, and suggest that the pellicle may serve as a model for engineered active surfaces with applications in microfluidics.

Shape dynamics, lipid hydrodynamics, and the complex viscoelasticity of bilayer membranes

María Jesús Samper — Wed, 26 Feb 2020 10:36:49 +0100

Biological membranes are continuously brought out of equilibrium, as they shape organelles, package and transport cargo, or respond to external actions. Even the dynamics of plain lipid membranes in experimental model systems are very complex due to the tight interplay between the bilayer architecture, the shape dynamics, and the rearrangement of the lipid molecules. We formulate and numerically implement a continuum model of the shape dynamics and lipid hydrodynamics, which describes the bilayer by its midsurface and by a lipid density field for each monolayer. The viscoelastic response of bilayers is determined by the stretching and curvature elasticity, and by the inter-monolayer friction and the membrane interfacial shear viscosity. While the bilayer equilibria are well understood theoretically, dynamical calculations have relied on simplified continuum approaches of uncertain transferability, or on molecular simulations reaching very limited length and time scales. Our approach incorporates the main physics, is fully nonlinear, does not assume predefined shapes, and can access a wide range of time and length scales. We validate it with the well understood tether extension. We investigate the tubular lipid transport between cells, the dynamics of bud absorption by a planar membrane, and the fate of a localized lipid density asymmetry in vesicles. These axisymmetric examples bear biological relevance and highlight the diversity of dynamical regimes that bilayers can experience.

Thin-shell theory based analysis of radially pressurized multiwall carbon nanotubes

María Jesús Samper — Wed, 26 Feb 2020 10:26:41 +0100

The radial deformation of multiwall carbon nanotubes (MWNTs) under hydrostatic pressure is investi gated within the continuum elastic approximation. A thin shell theory, with accurate elastic constants and interwall couplings, allows us to estimate the critical pressure above which the original circular cross section transforms into radially corrugated ones. The emphasis is placed on the rigorous formula tion of the van der Waals interaction between adjacent walls, which we analyze using two different approaches. Possible consequences of the radial corrugation in the physical properties of pressurized MWNTs are also discussed.

Thin shell analysis from scattered points with maximum-entropy approximants

María Jesús Samper — Wed, 26 Feb 2020 10:07:32 +0100

We present a method to process embedded smooth manifolds using sets of points alone. This method avoids any global parameterization and hence is applicable to surfaces of any genus. It combines three ingredients: (1) the automatic detection of the local geometric structure of the manifold by statistical learning methods; (2) the local parameterization of the surface using smooth meshfree (here maximum-entropy) approximants; and (3) patching together the local representations by means of a partition of unity. Mesh-based methods can deal with surfaces of complex topology, since they rely on the element-level parameterizations, but cannot handle high-dimensional manifolds, whereas previous meshfree methods for thin shells consider a global parametric domain, which seriously limits the kinds of surfaces that can be treated. We present the implementation of the method in the context of Kirchhoff–Love shells, but it is applicable to other calculations on manifolds in any dimension. With the smooth approximants, this fourth-order partial differential equation is treated directly. We show the good performance of the method on the basis of the classical obstacle course. Additional calculations exemplify the flexibility of the proposed approach in treating surfaces of complex topology and geometry.

On the optimum support size in meshfree methods: a variational adaptivity approach with maximum-entropy approximants

María Jesús Samper — Wed, 26 Feb 2020 09:50:33 +0100

We present a method for the automatic adaption of the support size of meshfree basis functions in the context of the numerical approximation of boundary value problems stemming from a minimum principle. The method is based on a variational approach, and the central idea is that the variational principle selects both the discretized physical fields and the discretization parameters, here those defining the support size of each basis function. We consider local maximum-entropy approximation schemes, which exhibit smooth basis functions with respect to both space and the discretization parameters (the node location and the locality parameters). We illustrate by the Poisson, linear and non-linear elasticity problems the effectivity of the method, which produces very accurate solutions with very coarse discretizations and finds unexpected patterns of the support size of the shape functions.

Diverse corrugation pattern in radially shrinking carbon nanotubes

María Jesús Samper — Tue, 25 Feb 2020 15:27:09 +0100

Stable cross sections of multiwalled carbon nanotubes subjected to electron-beam irradiation are investigated in the realm of the continuum mechanics approximation. The self-healing nature of sp2 graphitic sheets implies that selective irradiation of the outermost walls causes their radial shrinkage with the remaining inner walls undamaged. The shrinking walls exert high pressure on the interior part of nanotubes, yielding a wide variety of radial-corrugation patterns (i.e. circumferentially wrinkling structures) in the cross section. All corrugation patterns can be classified into two deformation phases for which the corrugation amplitudes of the innermost wall differ significantly.

Elastic bending modulus of monolayer graphene

María Jesús Samper — Tue, 25 Feb 2020 15:15:19 +0100

An analytic formula is derived for the elastic bending modulus of monolayer graphene based on an empirical potential for solid-state carbon atoms. Two physical origins are identified for the non-vanishing bending stiffness of the atomically thin graphene sheet, one due to the bond-angle effect and the other resulting from the bond-order term associated with the dihedral angles. The analytical prediction compares closely with ab initio energy calculations. Pure bending of graphene monolayers into cylindrical tubes is simulated by a molecular mechanics approach, showing slight nonlinearity and anisotropy in the tangent bending modulus as the bending curvature increases. An intrinsic coupling between bending and in-plane strain is noted for graphene monolayers rolled into carbon nanotubes.

Smooth, second order, non-negative meshfree approximants selected by maximum entropy

María Jesús Samper — Tue, 25 Feb 2020 15:09:33 +0100

We present a family of approximation schemes, which we refer to as second-order maximum-entropy (max-ent) approximation schemes, that extends the first-order local max-ent approximation schemes to second-order consistency. This method retains the fundamental properties of first-order max-ent schemes, namely the shape functions are smooth, non-negative, and satisfy a weak Kronecker-delta property at the boundary. This last property makes the imposition of essential boundary conditions in the numerical solution of partial differential equations trivial. The evaluation of the shape functions is not explicit, but it is very efficient and robust. To our knowledge, the proposed method is the first higher-order scheme for function approximation from unstructured data in arbitrary dimensions with non-negative shape functions. As a consequence, the approximants exhibit variation diminishing properties, as well as an excellent behavior in structural vibrations problems as compared with the Lagrange finite elements, MLS-based meshfree methods and even B-Spline approximations, as shown through numerical experiments. When compared with usual MLS-based second-order meshfree methods, the shape functions presented here are much easier to integrate in a Galerkin approach, as illustrated by the standard benchmark problems.

Relaxation dynamics of fluid membranes

María Jesús Samper — Tue, 25 Feb 2020 15:03:33 +0100

We study the effect of membrane viscosity in the dynamics of liquid membranes{possibly with free or internal boundaries{ driven by conservative forces (curvature elasticity and line tension) and dragged by the bulk dissipation of the ambient fluid and the friction occurring when the amphiphilic molecules move relative to each other. To this end, we formulate a continuum model which includes a new form of the governing equations for a two-dimensional viscous fluid moving on a curved, time-evolving surface. The effect of membrane viscosity has received very limited attention in previous continuum studies of the dynamics of fluid membranes, although recent coarse-grained discrete simulations suggest its importance. By applying our model to the study of vesiculation and membrane fusion in a simpli ed geometry, we conclude that membrane viscosity plays a dominant role in the relaxation dynamics of fluid membranes of sizes comparable to those found in eukaryotic cells, and is not negligible in many large synthetic systems of current interest.

Effective coarse-grained simulations of super-thick multi-walled carbon nanotubes under torsion

María Jesús Samper — Tue, 25 Feb 2020 14:57:27 +0100

Under torsion and beyond the buckling point, multi-walled carbon nanotubes (MWCNTs) develop a periodic wave-like rippling morphology. Here, we show that torsional rippling deformations can be accurately described by a simple sinusoidal shape function. Combining this observation with the geometry optimization, we develop an effective coarse-grained model that reproduces the complex nonlinear mechanical responses of thick MWCNTs under torsion predicted by large-scale atomistic simulations. Furthermore, the model allows us to simulate super-thick tubes, inaccessible by other coarse-grained methods. With this effective coarse-grained model, we show from an energetic analysis that the rippling deformation is a result of in-plane strain energy relaxation, penalized by the increase in the interlayer van der Waals interaction energy. Our simulations reveal that the torsional response of MWCNTs with up to 100 layers approximately follows a simple bilinear law, and the ratio of the torsional rigidities in the pre- and post-buckling regimes is nearly a constant, independent of the tube radius. In contrast, the bifurcation torsional strain powerly scales with the tube radius. We also find that the wave number in the circumferential direction linearly increases with tube radius, while the wavelength monotonically increases with tube radius, and approaches a constant in the limit of bulk graphite. The bilinear constitutive relation, together with the scaling law of the bifurcation torsional strain, furnishes a simple nonlinear beam theory, which facilitates the analysis of MWCNT bundles and networks.

Optimization of a novel micro-opto-X-ray imaging lens

María Jesús Samper — Tue, 25 Feb 2020 14:48:53 +0100

A mechanically deformable reflection transmission MOEMS system is capable of focusing X-rays to sub micron spots. This paper considers the geometry of the proposed deformed slotted microcantilever lens element under thermally derived strain, using finite element analysis. The work shows that an optimized MOEMS design using a slotted polyimide/gold thermal bimorph cantilever is capable of achieving ideal geometry with a larger than expected number of focusing slots – up to 111 at 2 μm width.

Rippling and a phase-transforming mesoscopic model for multiwalled carbon nanotubes

María Jesús Samper — Tue, 25 Feb 2020 14:31:40 +0100

We propose to model thick multiwalled carbon nanotubes as beams with non-convex curvature energy. Such models develop stressed phase mixtures composed of smoothly bent sections and rippled sections. This model is motivated by experimental observations and large-scale atomistic-based simulations. The model is analyzed, validated against large-scale simulations, and exercised in examples of interest. It is shown that modelling MWCNTs as linear elastic beams can result in poor approximations that overestimate the elastic restoring force considerably, particularly for thick tubes. In contrast, the proposed model produces very accurate predictions both of the restoring force and of the phase pattern. The size effect in the bending response of MWCNTs is also discussed.

Size-dependent nonlinear elastic scaling of multiwalled carbon nanotubes

María Jesús Samper — Tue, 25 Feb 2020 14:25:14 +0100

We characterize through large-scale simulations the nonlinear elastic response of multi-walled carbon nanotubes (MWNCNTs) in torsion and bending. We identify a unified law consisting of two distinct power-law regimes in the energy-deformation relation. This law encapsulates the complex mechanics of rippling and is described in terms of elastic constants, a critical length-scale and an anharmonic energy-deformation exponent. The mechanical response of MWCNTs is found to be strongly size-dependent, in that the critical strain beyond which they behave nonlinearly scales as the inverse of their diameter. These predictions are consistent with available experimental observations.

List of publications of Aula CIMNE UPCT

Scipedia content — Tue, 25 Feb 2020 10:09:42 +0100

List of publications of Aula CIMNE UPCT

R&D Projects of Aula CIMNE UPCT

Scipedia content — Tue, 25 Feb 2020 10:01:52 +0100

R&D Projects of Aula CIMNE UPCT

On the geometric character of stress in continuum mechanics

María Jesús Samper — Tue, 25 Feb 2020 09:56:07 +0100

This paper shows that the stress field in the classical theory of continuum mechanics may be taken to be a covector-valued differential two-form. The balance laws and other fundamental laws of continuum mechanics may be neatly rewritten in terms of this geometric stress. A geometrically attractive and covariant derivation of the balance laws from the principle of energy balance in terms of this stress is presented.

Prueba

Héctor Tabares-Ospina — Sat, 22 Feb 2020 23:30:04 +0100

This

Active vibration control analysis in smart composite structures using ANSYS

AbdulRahman B. Shakir — Fri, 21 Feb 2020 22:55:03 +0100

Piezoelectric Macro-Fiber Composite (MFC) utilization is increasing in engineering fields due to its strong actuation forces and high flexibility. In this paper, piezoelectric ( ) type (M8528-P1) patches are applied for active control of a cantilever composite beam. A linear coupled ﬁnite element model for piezoelectric MFC actuation of the composite beam was developed based on the APDL-ANSYS codes by using linear piezoelectric constitutive equations to study smart composite beam behavior in open and closed-loop cases. Active vibration control response of laminated composite beams with various stacking sequence configurations was examined, and the results were compared. In this, the proportional type (kp) of control algorithms is utilized, and in the summary of work refined two sets of finite element models. The first set their laminates to have an orientation (0 with 90), and others have an orientation (0 with 45). When the control signal applied with the gain (kp) to a system increases, the rise time generally decreases, the actuator voltages for different control gain for all cases are observed with the increase in the proportional constant. The findings of this study indicate that the composite beams composed from (0/90) stacking sequence sets had reached stability state faster than the composite beams composed from (0/45) stacking sequence sets.

Local maximum‐entropy approximation schemes: a seamless bridge between finite elements and meshfree methods

María Jesús Samper — Fri, 21 Feb 2020 15:05:20 +0100

We present a one‐parameter family of approximation schemes, which we refer to as local maximum‐entropy approximation schemes, that bridges continuously two important limits: Delaunay triangulation and maximum‐entropy (max‐ent) statistical inference. Local max‐ent approximation schemes represent a compromise—in the sense of Pareto optimality—between the competing objectives of unbiased statistical inference from the nodal data and the definition of local shape functions of least width. Local max‐ent approximation schemes are entirely defined by the node set and the domain of analysis, and the shape functions are positive, interpolate affine functions exactly, and have a weak Kronecker‐delta property at the boundary. Local max‐ent approximation may be regarded as a regularization, or thermalization, of Delaunay triangulation which effectively resolves the degenerate cases resulting from the lack or uniqueness of the triangulation. Local max‐ent approximation schemes can be taken as a convenient basis for the numerical solution of PDEs in the style of meshfree Galerkin methods. In test cases characterized by smooth solutions we find that the accuracy of local max‐ent approximation schemes is vastly superior to that of finite elements.

Continuum mechanics modeling and simulation of carbon nanotubes

María Jesús Samper — Fri, 21 Feb 2020 14:51:02 +0100

The understanding of the mechanics of atomistic systems greatly benefits from continuum mechanics. One appealing approach aims at deductively constructing continuum theories starting from models of the interatomic interactions. This viewpoint has become extremely popular with the quasicontinuum method. The application of these ideas to carbon nanotubes presents a peculiarity with respect to usual crystalline materials: their structure relies on a two-dimensional curved lattice. This renders the cornerstone of crystal elasticity, the Cauchy–Born rule, insufficient to describe the effect of curvature. We discuss the application of a theory which corrects this deficiency to the mechanics of carbon nanotubes (CNTs). We review recent developments of this theory, which include the study of the convergence characteristics of the proposed continuum models to the parent atomistic models, as well as large scale simulations based on this theory. The latter have unveiled the complex nonlinear elastic response of thick multiwalled carbon nanotubes (MWCNTs), with an anomalous elastic regime following an almost absent harmonic range.

Finite crystal elasticity of carbon nanotubes based on the exponential Cauchy-Born rule

María Jesús Samper — Fri, 21 Feb 2020 14:35:58 +0100

A finite deformation continuum theory is derived from interatomic potentials for the analysis of the mechanics of carbon nanotubes. This nonlinear elastic theory is based on an extension of the Cauchy-Born rule called the exponential Cauchy-Born rule. The continuum object replacing the graphene sheet is a surface without thickness. The method systematically addresses both the characterization of the small strain elasticity of nanotubes and the simulation at large strains. Elastic moduli are explicitly expressed in terms of the functional form of the interatomic potential. The expression for the flexural stiffness of graphene sheets, which cannot be obtained from standard crystal elasticity, is derived. We also show that simulations with the continuum model combined with the finite element method agree very well with zero temperature atomistic calculations involving severe deformations.

Finite element methods for the non-linear mechanics of crystalline sheets and nanotubes

María Jesús Samper — Fri, 21 Feb 2020 14:18:06 +0100

The formulation and finite element implementation of a finite deformation continuum theory for the mechanics of crystalline sheets is described. This theory generalizes standard crystal elasticity to curved monolayer lattices by means of the exponential Cauchy–Born rule. The constitutive model for a two‐dimensional continuum deforming in three dimensions (a surface) is written explicitly in terms of the underlying atomistic model. The resulting hyper‐elastic potential depends on the stretch and the curvature of the surface, as well as on internal elastic variables describing the rearrangements of the crystal within the unit cell. Coarse grained calculations of carbon nanotubes (CNTs) are performed by discretizing this continuum mechanics theory by finite elements. A smooth discrete representation of the surface is required, and subdivision finite elements, proposed for thin‐shell analysis, are used. A detailed set of numerical experiments, in which the continuum/finite element solutions are compared to the corresponding full atomistic calculations of CNTs, involving very large deformations and geometric instabilities, demonstrates the accuracy of the proposed approach. Simulations for large multi‐million systems illustrate the computational savings which can be achieved

Nonlinear mechanical response and rippling of thick multi-walled carbon nanotubes

María Jesús Samper — Fri, 21 Feb 2020 14:02:18 +0100

The measured drop of the effective bending stiffness of multiwalled carbon nanotubes (MWCNTs) with increasing diameter is investigated by a generalized local quasicontinuum method. The previous hypothesis that this reduction is due to a rippling mode is confirmed by the calculations. The observed ripples result from a complex three-dimensional deformation similar to theYoshimura buckling pattern. It is found that thick MWCNTs exhibit a well-defined nonlinear moment-curvature relation, even for small deformations, governed by the interplay of strain energy relaxation and intertube interactions. Rippling deformations are also predicted for MWCNTs subject to torsion, resulting in an effective torsional modulus much smaller than that predicted by linear elasticity.

A finite deformation membrane based on inter-atomic potentials for the transverse mechanics of nanotubes

María Jesús Samper — Fri, 21 Feb 2020 13:49:21 +0100

A finite deformation hyper-elastic membrane theory based on inter-atomic potentials for crystalline films composed of a single atomic layer is developed. For this purpose, an extension of the standard Born rule that exploits the differential geometry concept of the exponential map is proposed to deal with the curvature of surfaces. The exponential map is approximated locally and strain measures based on the stretch and the curvature of the membrane arise. The methodology is first particularized to atomic chains in two dimensions, and then to graphene sheets. A reduced model for the transverse mechanics of carbon nanotubes is developed in detail. This model is a hyper-elastic constrained membrane which fully exploits the symmetry of the transverse deformation. Additionally, a continuum version of the non-bonded interactions is provided. The continuum model is discretized using finite elements and very good agreement with molecular mechanics simulations is obtained. Finally, several simulations illustrate the strong effect of the van der Waals interactions in the transverse deformation of carbon nanotubes.

An atomistic-based finite deformation membrane for single layer crystalline films

María Jesús Samper — Fri, 21 Feb 2020 13:39:55 +0100

A general methodology to develop hyper-elastic membrane models applicable to crystalline films one-atom thick is presented. In this method, an extension of the Born rule based on the exponential map is proposed. The exponential map accounts for the fact that the lattice vectors of the crystal lie along the chords of the curved membrane, and consequently a tangent map like the standard Born rule is inadequate. In order to obtain practical methods, the exponential map is locally approximated. The effectiveness of our approach is demonstrated by numerical studies of carbon nanotubes. Deformed configurations as well as equilibrium energies of atomistic simulations are compared with those provided by the continuum membrane resulting from this method discretized by finite elements.

A generic finite element framework on parallel tree-based adaptive meshes

Cecilia Soriano — Fri, 21 Feb 2020 11:43:02 +0100

We present highly scalable parallel distributed-memory algorithms and associated data structures for a generic finite element framework that supports h-adaptivity on computational domains represented as multiple connected adaptive trees—forest-of-trees—, thus providing multi-scale resolution on problems governed by partial differential equations.The framework is grounded on a rich representation of the adaptive mesh suitable for generic finite elements that is built on top of a low-level, light-weight forest-oftrees data structure handled by a specialized, highly parallel adaptive meshing engine. Along the way, we have identified the requirements that the forest-of-trees layer must fulfill to be coupled into our framework. Essentially, it must be able to describe neighboring relationships between cells in the adapted mesh (apart from hierarchical relationships) across the lower-dimensional objects at the boundary of the cells. Atop this two-layered mesh representation, we build the rest of data structures required for the numerical integration and assembly of the discrete system of linear equations.We consider algorithms that are suitable for both subassembled and fully-assembled distributed data layouts of linear system matrices. The proposed framework has been implemented within the FEMPAR scientific software library, using p4est as a practical forest-of-octrees demonstrator. A comprehensive strong scaling study of this implementation when applied to Poisson and Maxwell problems reveals remarkable scalability up to 32.2K CPU cores and 482.2M degrees of freedom. Besides, the implementation in FEMPAR of the proposed approach is up to 2.6 and 3.4 times faster than the state-of-the-art deal.II finite element software in the h-adaptive approximation of a Poisson problem with firstand second-order Lagrangian finite elements, respectively (excluding the linear solver step from the comparison).

Embedded multilevel Monte Carlo for uncertainty quantification in random domains

Cecilia Soriano — Fri, 21 Feb 2020 11:36:03 +0100

The multilevel Monte Carlo (MLMC) method has proven to be an effective variance-reduction statistical method for Uncertainty quantification in PDE models. It combines approximations at different levels of accuracy using a hierarchy of meshes in a similar way as multigrid. The generation of body-fitted mesh hierarchies is only possible for simple geometries. On top of that, MLMC for random domains involves the generation of a mesh for every sample. Instead, here we consider the use of embedded methods which make use of simple background meshes of an artificial domain (a bounding-box) for which it is easy to define a mesh hierarchy, thus eliminating the need of body-fitted unstructured meshes, but can produce ill-conditioned discrete problems. To avoid this complication, we consider the recent aggregated finite element method (AgFEM). In particular, we design an embedded MLMC framework for (geometrically and topologically) random domains implicitly defined through a random level-set function, which makes use of a set of hierarchical background meshes and the AgFEM. Performance predictions from existing theory are verified statistically in three numerical experiments, namely the solution of the Poisson equation on a circular domain of random radius, the solution of the Poisson equation on a topologically identical but more complex domain, and the solution of a heat-transfer problem in a domain that has geometric and topological uncertainties. Finally, the use of AgFE is statistically demonstrated to be crucial for complex and uncertain geometries in terms of robustness and computational cost. Date: November 28, 2019.

A modified Finite Element formulation for the imposition of the slip boundary condition over embedded volumeless geometries

Cecilia Soriano — Fri, 21 Feb 2020 11:33:02 +0100

This work describes a novel formulation for the simulation of Navier-Stokes problems including embedded objects. The new proposal is based on the use of a modified finite element space, which replaces the standard one within the elements intersected by the immersed geometry. The modified space is able to exactly reproduce the jumps happening at the embedded boundary while preserving the conformity across the faces intersected by the embedded object. The paper focuses particularly on the imposition of a slip boundary condition on the surface of the embedded geometry, proposing a new technique for the application of such constraint. The new proposal is carefully benchmarked using the results of a body fitted technique and of an alternative embedded approach. Potential applications of interest are also presented.

Computational modeling of the fluid flow in type B aortic dissection using a modified finite element embedded formulation

Cecilia Soriano — Fri, 21 Feb 2020 11:25:16 +0100

This work explores the use of an embedded computational fluid dynamics method to study the type B aortic dissection. The use of the proposed technique makes it possible to easily test different intimal flap configurations without any need of remeshing. To validate the presented methodology, we take as reference test case an in vitro experiment present in the literature. This experiment, which considers several intimal flap tear configurations (number, size and location), mimics the blood flow in a real type B aortic dissection. We prove the correctness and suitability of the presented approach by comparing the pressure values and waveform. The obtained results exhibit a remarkable similarity with the experimental reference data. Complementary, we present a feasible surgical application of the presented computer method. The aim is to help the clinicians in the decision making before the type B aortic dissection surgical fenestration. The capabilities of the proposed technique are exploited to efficiently create artificial reentry tear configurations. We highlight that only the radius and center of the reentry tear need to be specified by the clinicians, without any need to modify neither the model geometry nor the mesh. The obtained computational surgical fenestration results are in line with the medical observations in similar clinical studies.

Parallel unstructured mesh adaptation using iterative remeshing and repartitioning

Cecilia Soriano — Fri, 21 Feb 2020 11:04:35 +0100

Mesh adaptation has proven to be a powerful tool for increasing the accuracy ofnumerical simulations whenever the solution exhibits strong non-uniform features over the com-putational domain. Sequential remeshing currently represents a bottleneck for parallel solvers. Tothis aim, we present a parallel remeshing algorithm that enables the reuse of existing sequential remeshing libraries, a non-intrusive linkage with third-party solvers, and an efficient exploitationof distributed parallel environments. The numerical procedure is implemented in the open sourceParMmg software package.

Balancing domain decomposition by constraints associated with subobjects

Cecilia Soriano — Fri, 21 Feb 2020 10:55:02 +0100

A simple variant of the BDDC preconditioner in which constraints are imposed on a selected set of subobjects (subdomain subedges, subfaces and vertices between pairs of subedges) is presented. We are able to show that the condition number of the preconditioner is bounded by C 1 + log(L/h)2, where C is a constant, and h and L are the characteristic sizes of the mesh and the subobjects, respectively. As L can be chosen almost freely, the condition number can theoretically be as small as O(1). We will discuss the pros and cons of the preconditioner and its application to heterogeneous problems. Numerical results on supercomputers are provided.

Distributed-memory parallelization of the aggregated unfitted finite element method

Cecilia Soriano — Fri, 21 Feb 2020 10:47:03 +0100

The aggregated unfitted finite element method (AgFEM) is a methodology recently introduced in order to address conditioning and stability problems associated with embedded, unfitted, or extended finite element methods. The method is based on removal of basis functions associated with badly cut cells by introducing carefully designed constraints, which results in well-posed systems of linear algebraic equations, while preserving the optimal approximation order of the underlying finite element spaces. The specific goal of this work is to present the implementation and performance of the method on distributed-memory platforms aiming at the efficient solution of large-scale problems. In particular, we show that, by considering AgFEM, the resulting systems of linear algebraic equations can be effectively solved using standard algebraic multigrid preconditioners. This is in contrast with previous works that consider highly customized preconditioners in order to allow one the usage of iterative solvers in combination with unfitted techniques. Another novelty with respect to the methods available in the literature is the problem sizes that can be handled with the proposed approach. While most of previous references discussing linear solvers for unfitted methods are based on serial non-scalable algorithms, we propose a parallel distributed-memory method able to efficiently solve problems at large scales. This is demonstrated by means of a weak scaling test defined on complex 3D domains up to 300M degrees of freedom and one billion cells on 16K CPU cores in the Marenostrum-IV platform. The parallel implementation of the AgFEM method is available in the large-scale finite element package FEMPAR.

The surrogate matrix methodology: A reference implementation for low-cost assembly in isogeometric analysis

Cecilia Soriano — Fri, 21 Feb 2020 10:41:02 +0100

A reference implementation of a new method in isogeometric analysis (IGA) is presented. It delivers lowcost variable-scale approximations (surrogates) of the matrices which IGA conventionally requires to be computed by element-scale quadrature. To generate surrogate matrices, quadrature must only be performed on a fraction of the elements in the computational domain. In this way, quadrature determines only a subset of the entries in the final matrix. The remaining matrix entries are computed by a simple B-spline interpolation procedure. We present the modifications and extensions required for a reference implementation in the open-source IGA software library GeoPDEs. The exposition is fashioned to help facilitate similar modifications in other contemporary software libraries. Method name: Surrogate matrix method for isogeometric analysis

The surrogate matrix methodology: Low-cost assembly for isogeometric analysis

Cecilia Soriano — Fri, 21 Feb 2020 10:18:03 +0100

A new methodology in isogeometric analysis (IGA) is presented. This methodology delivers low-cost variable-scale approximations (surrogates) of the matrices which IGA conventionally requires to be computed from element-scale quadrature formulas. To generate surrogate matrices, quadrature must only be performed on certain elements in the computational domain. This, in turn, determines only a subset of the entries in the final matrix.The remaining matrix entries are computed by a simple B-spline interpolation procedure. Poisson’s equation, membrane vibration, plate bending, and Stokes’ flow problems are studied. In these problems, the use of surrogate matrices has a negligible impact on solution accuracy. Because only a small fraction of the original quadrature must be performed, we are able to report beyond a fifty-fold reduction in overall assembly time in the same software. The capacity for even further speed-ups is clearly demonstrated. The implementation used here was achieved by a small number of modifications to the open-source IGA software library GeoPDEs. Similar modifications could be made to other present-day software libraries.

The surrogate matrix methodology: a priori error estimation

Cecilia Soriano — Fri, 21 Feb 2020 09:59:22 +0100

We give the first mathematically rigorous analysis of an emerging approach to finite element analysis (see, e.g., Bauer et al. [Appl. Numer. Math., 2017]), which we hereby refer to as the surrogate matrix methodology. This methodology is based on the piece-wise smooth approximation of the matrices involved in a standard finite element discretization. In particular, it relies on the projection of smooth so-called stencil functions onto high-order polynomial subspaces. The performance advantage of the surrogate matrix methodology is seen in constructions where each stencil function uniquely determines the values of a significant collection of matrix entries. Such constructions are shown to be widely achievable through the use of locally-structured meshes. Therefore, this methodology can be applied to a wide variety of physically meaningful problems, including nonlinear problems and problems with curvilinear geometries. Rigorous a priori error analysis certifies the convergence of a novel surrogate method for the variable coefficient Poisson equation. The flexibility of the methodology is also demonstrated through the construction of novel methods for linear elasticity and nonlinear diffusion problems. In numerous numerical experiments, we demonstrate the efficacy of these new methods in a matrix-free environment with geometric multigrid solvers. In our experiments, up to a twenty-fold decrease in computation time is witnessed over the classical method with an otherwise identical implementation.

AutoParallel: A Python module for automatic parallelization and distributed execution of affine loop nests

Cecilia Soriano — Fri, 21 Feb 2020 09:42:37 +0100

The last improvements in programming languages, programming models, and frameworks have focused on abstracting the users from many programming issues. Among others, recent programming frameworks include simpler syntax, automatic memory management and garbage collection, which simplifies code re-usage through library packages, and easily configurable tools for deployment. For instance, Python has risen to the top of the list of the programming languages due to the simplicity of its syntax, while still achieving a good performance even being an interpreted language. Moreover, the community has helped to develop a large number of libraries and modules, tuning them to obtain great performance.
However, there is still room for improvement when preventing users from dealing directly with distributed and parallel computing issues. This paper proposes and evaluates AutoParallel, a Python module to automatically find an appropriate task-based parallelization of affine loop nests to execute them in parallel in a distributed computing infrastructure. This parallelization can also include the building of data blocks to increase task granularity in order to achieve a good execution performance. Moreover, AutoParallel is based on sequential programming and only contains a small annotation in the form of a Python decorator so that anyone with little programming skills can scale up an application to hundreds of cores.

Quantitative investment model based on data mining

Daxing Wang — Wed, 19 Feb 2020 12:45:48 +0100

Quantitative investment is the process of establishing mathematical models using statistics, information technology, and mathematics to quantify and implement risks, returns, and traditional investment concepts. However, due to the backwardness of computing tools in the past, quantitative investment has not received much recognition. With the improvement of computer science and quantitative analysis theory, traditional fundamental analysis and the use of sampling statistical technology to build advanced mathematical models for investment analysis have failed to meet the requirements of investors. Therefore, the Quantitative investment strategies based on data mining technology are receiving more and more attention. In this paper, we uses MATLAB software to capture big data from financial and economic websites, and then uses neural network training models to predict the trend of stock changes, and finally establishes a suitable quantitative stock selection model. The simulation results show that only by using quantitative stock selection strategies to curb risks and selecting a suitable investment portfolio can achieve the ideal goals in the stock market.

Explicit finite deformation stress integration of the elasto-plastic constitutive equations

María Jesús Samper — Tue, 18 Feb 2020 15:07:44 +0100

In this work, an explicit integration scheme for hyperelastic-based finite strains elasto-plasticity is presented. One step update equations are obtained from the large deformation multiplicative elasto-plasticity theory, where an exponential variation of the plastic deformation gradient is assumed. In addition, the material tangent matrix is presented and has the same formal structure as the usual small strains elasto-plastic tangent matrix. In purely elastic regime, the proposal reduces to the usual large deformation elasticity equations. Routines to alleviate the main drawbacks of explicit methods are outlined, such as a yield surface drift correction scheme and an adaptive substepping method. Several examples of typical geotechnical tests using the Houlsby hyperelastic model along with the Modified Cam Clay plastic model are discussed. Results from a convergence test suggest that, using an adaptive substepping scheme, the error of the local problem is independent of the step size.

Evaluation of Multilayer Pavement Viscoelastic Properties from Falling Weight Deflectometer using Neural Networks

María Jesús Samper — Tue, 18 Feb 2020 13:56:19 +0100

The surface deflection caused by the falling weight deflectometer (FWD) is a dynamic problem usually treated as a static problem. This chapter presents a dynamic solution of the backcalculation problem using a viscoelastic model, introducing a viscosity variable. The calibration process is based on the numerical simulation of the FWD test to approximate the deflection curves obtained in the road track. Description of the experimental curves and the hypothesis of the numerical problem assumed are also described. The samples used to train and validate the artificial neural network (ANN) are conformed by four values per curve times the number of geophones, the input data, while the output data are five Young moduli and five viscosities. The backcalculation process is based on experimental tests from the FWD and calibrated using an FEM simulation of the problem.

The particle finite element method (PFEM) in thermo‐mechanical problems

María Jesús Samper — Tue, 18 Feb 2020 13:44:27 +0100

The aim of this work is to develop a numerical framework for accurately and robustly simulating the different conditions exhibited by thermo‐mechanical problems. In particular, the work will focus on the analysis of problems involving large strains, rotations, multiple contacts, large boundary surface changes, and thermal effects.

The framework of the numerical scheme is based on the particle finite element method (PFEM) in which the spatial domain is continuously redefined by a distinct nodal reconnection, generated by a Delaunay triangulation. In contrast to classical PFEM calculations, in which the free boundary is obtained by a geometrical procedure (α − shape method), in this work, the boundary is considered as a material surface, and the boundary nodes are removed or inserted by means of an error function.

The description of the thermo‐mechanical constitutive model is based on the concepts of large strains plasticity. The plastic flow condition is assumed nearly incompressible, so a u‐p mixed formulation, with a stabilization of the pressure term via the polynomial pressure projection, is proposed.

One of the novelties of this work is the use of a combination between the isothermal split and the so‐called IMPL‐EX hybrid integration technique to enhance the robustness and reduce the typical iteration number of the fully implicit Newton–Raphson solution algorithm.

The new set of numerical tools implemented in the PFEM algorithm, including new discretization techniques, the use of a projection of the variables between meshes, and the insertion and removal of points allows us to eliminate the negative Jacobians present during large deformation problems, which is one of the drawbacks in the simulation of coupled thermo‐mechanical problems.

Finally, two sets of numerical results in 2D are stated. In the first one, the behavior of the proposed locking‐free element type and different time integration schemes for thermo‐mechanical problems is analyzed. The potential of the method for modeling more complex coupled problems as metal cutting and metal forming processes is explored in the last example.

Numerical simulation of undrained insertion problems in geotechnical engineering with the Particle Finite Element Method (PFEM)

María Jesús Samper — Tue, 18 Feb 2020 13:21:25 +0100

The paper presents total-stress numerical analyses of large-displacement soil-structure interaction problems in geomechanics using the Particle Finite Element Method (PFEM). This method is characterized by frequent remeshing and the use of low order finite elements to evaluate the solution. Several important features of the method are: (i) a mixed formulation (displacement-mean pressure) stabilized numerically to alleviate the volumetric locking effects that are characteristic of low order elements when the medium is incompressible, (ii) a penalty method to prescribe the contact constraints between a rigid body and a deformable media combined with an implicit scheme to solve the tangential contact constraint, (iii) an explicit algorithm with adaptive substepping and correction of the yield surface drift to integrate the finite-strain multiplicative elasto-plastic constitutive relationship, and (iv) the mapping schemes to transfer information between successive discretizations. The performance of the method is demonstrated by several numerical examples, of increasing complexity, ranging from the insertion of a rigid strip footing to a rough cone penetration test. It is shown that the proposed method requires fewer computational resources than other numerical approaches addressing the same type of problems.

Damage analysis of masonry structures subjected to rockfalls

María Jesús Samper — Tue, 18 Feb 2020 12:50:26 +0100

Masonry structures present substantial vulnerability to rockfalls. The methodologies for the damage quantification of masonry structures subjected to rockfalls are scarce. An analytical procedure for the damage assessment of masonry structures is presented. The procedure comprises three stages: (1) determination of the rockfall impact actions which are applied to a masonry structure, in terms of external forces, using the particle finite element method (PFEM), (2) evaluation of the mechanical properties, modelling of the masonry structure, and calculation of the internal stresses, using the finite element method (FEM), (3) assessment of the damage due to the rockfall actions, applying a failure criterion adapted to masonries, and calculation of the damage in terms of the percentage of the damaged wall surface. Three real rockfall events and their impact on buildings are analysed. A sensitivity analysis of the proposed procedure is then used to identify the variables that mostly affect the extent of the wall damage, which are the masonry width, the tensile strength, the block diameter and lastly, velocity.

Performance of mixed formulations for the particle finite element method in soil mechanics problems

María Jesús Samper — Tue, 18 Feb 2020 11:52:14 +0100

This paper presents a computational framework for the numerical analysis of fluid-saturated porous media at large strains. The proposal relies, on one hand, on the particle finite element method (PFEM), known for its capability to tackle large deformations and rapid changing boundaries, and, on the other hand, on constitutive descriptions well established in current geotechnical analyses (Darcy’s law; Modified Cam Clay; Houlsby hyperelasticity). An important feature of this kind of problem is that incompressibility may arise either from undrained conditions or as a consequence of material behaviour; incompressibility may lead to volumetric locking of the low-order elements that are typically used in PFEM. In this work, two different three-field mixed formulations for the coupled hydromechanical problem are presented, in which either the effective pressure or the Jacobian are considered as nodal variables, in addition to the solid skeleton displacement and water pressure. Additionally, several mixed formulations are described for the simplified single-phase problem due to its formal similitude to the poromechanical case and its relevance in geotechnics, since it may approximate the saturated soil behaviour under undrained conditions. In order to use equal-order interpolants in displacements and scalar fields, stabilization techniques are used in the mass conservation equation of the biphasic medium and in the rest of scalar equations. Finally, all mixed formulations are assessed in some benchmark problems and their performances are compared. It is found that mixed formulations that have the Jacobian as a nodal variable perform better.

Continuous chip formation in metal cutting processes using the Particle Finite Element Method (PFEM)

María Jesús Samper — Tue, 18 Feb 2020 11:38:28 +0100

This paper presents a study on the metal cutting simulation with a particular numerical technique, the Particle Finite Element Method (PFEM) with a new modified time integration algorithm and incorporating a contact algorithm capability . The goal is to reproduce the formation of continuous chip in orthogonal machining. The paper tells how metal cutting processes can be modelled with the PFEM and which new tools have been developed to provide the proper capabilities for a successful modelling. The developed method allows for the treatment of large deformations and heat conduction, workpiece-tool contact including friction effects as well as the full thermo-mechanical coupling for contact. The difficulties associated with the distortion of the mesh in areas with high deformation are solved introducing new improvements in the continuous Delaunay triangulation of the particles. The employment of adaptative insertion and removal of particles at every new updated configuration improves the mesh quality allowing for resolution of finer-scale features of the solution. The performance of the method is studied with a set of different two-dimensional tests of orthogonal machining. The examples consider, from the most simple case to the most complex case, different assumptions for the cutting conditions and different material properties. The results have been compared with experimental tests showing a good competitiveness of the PFEM in comparison with other available simulation tools.

Coronal advanced flap in combination with a connective tissue graft. Is the thickness of the flap a predictor for root coverage? A prospective clinical study

María Jesús Samper — Tue, 18 Feb 2020 09:55:56 +0100

Aim

Evaluate if there is any relationship between the flap thickness (FT) and the presence of complete root coverage (CRC) when performing coronally advanced flaps in combination with a connective tissue graft (CTG).

Materials and methods

Prospective clinical study, in which multiple Miller class I and II recessions were treated with a coronally advanced flap and a CTG standardized at 1 mm of thickness. Individual stents permitted repeated measurements of conventional periodontal parameters at the same point. The primary outcome variable was CRC. Secondary outcomes were recession reduction, gingival thickness and width of keratinized tissue (KT) achieved at 6 months post‐surgery.

Results

Forty‐five recessions (2.4 ± 0.75 mm) were treated in 20 patients. Mean root coverage was 93.4 ± 10.98%; 65% achieved CRC. The mean FT was 1.01 mm ± 0.64 mm and 1.01 ± 0.61 mm at 2 and 5 mm from the gingival margin, respectively. No relationship could be found between FT and CRC (p > .05). Statistical significant changes (p < .05) were observed for recession depth, clinical attachment level, KT and soft tissue thickness at the end of the study.

Conclusions

Flap thickness seems not to be a predictor for CRC when performing a coronally advanced flap plus a CTG. This technique may be of choice when treating thin biotypes.

Generation of segmental chips in metal cutting modeled with the PFEM

María Jesús Samper — Tue, 18 Feb 2020 09:32:28 +0100

The Particle Finite Element Method, a lagrangian finite element method based on a continuous Delaunay re-triangulation of the domain, is used to study machining of Ti6Al4V. In this work the method is revised and applied to study the influence of the cutting speed on the cutting force and the chip formation process. A parametric methodology for the detection and treatment of the rigid tool contact is presented. The adaptive insertion and removal of particles are developed and employed in order to sidestep the difficulties associated with mesh distortion, shear localization as well as for resolving the fine-scale features of the solution. The performance of PFEM is studied with a set of different two-dimensional orthogonal cutting tests. It is shown that, despite its Lagrangian nature, the proposed combined finite element-particle method is well suited for large deformation metal cutting problems with continuous chip and serrated chip formation.

FEMPAR: an object-oriented parallel finite element framework

María Jesús Samper — Mon, 17 Feb 2020 15:32:58 +0100

FEMPAR is an open source object oriented Fortran200X scientific software library for the high-performance scalable simulation of complex multiphysics problems governed by partial differential equations at large scales, by exploiting state-of-the-art supercomputing resources. It is a highly modularized, flexible, and extensible library, that provides a set of modules that can be combined to carry out the different steps of the simulation pipeline. FEMPAR includes a rich set of algorithms for the discretization step, namely (arbitrary-order) grad, div, and curl-conforming finite element methods, discontinuous Galerkin methods, B-splines, and unfitted finite element techniques on cut cells, combined with h-adaptivity. The linear solver module relies on state-of-the-art bulk-asynchronous implementations of multilevel domain decomposition solvers for the different discretization alternatives and block-preconditioning techniques for multiphysics problems. FEMPAR is a framework that provides users with out-of-the-box state-of-the-art discretization techniques and highly scalable solvers for the simulation of complex applications, hiding the dramatic complexity of the underlying algorithms. But it is also a framework for researchers that want to experience with new algorithms and solvers, by providing a highly extensible framework. In this work, the first one in a series of articles about FEMPAR, we provide a detailed introduction to the software abstractions used in the discretization module and the related geometrical module. We also provide some ingredients about the assembly of linear systems arising from finite element discretizations, but the software design of complex scalable multilevel solvers is postponed to a subsequent work.

Driving mechanisms and streamwise homogeneity in molecular dynamics simulations of nanochannel flows

María Jesús Samper — Mon, 17 Feb 2020 15:22:13 +0100

In molecular dynamics simulations, nanochannel flows are usually driven by a constant force, that aims to represent a pressure difference between inlet and outlet, and periodic boundary conditions are applied in the streamwise direction resulting in an homogeneous flow. The homogeneity hypothesis can be eliminated adding reservoirs at the inlet and outlet of the channel which permits us to predict streamwise variation of flow properties. It also opens the door to drive the flow by applying a pressure gradient instead of a constant force. We analyze the impact of these modeling modifications in the prediction of the flow properties, and we show when they make a difference with respect to the standard approach. It turns out that both assumptions are irrelevant when low pressure differences are considered, but important differences are observed at high pressure differences. They include the density and velocity variation along the channel (the mass flow rate is constant) but, more importantly, the temperature increase and slip length decrease. Because viscous heating is important at high shear rates, these modeling issues are also linked to the use of thermostating procedures. Specifically, selecting the region where the thermostat is applied has a critical influence on the results. Whereas in the traditional homogeneous model the choices are limited to the fluid and/or the wall, in the inhomogeneous cases the reservoirs are also available, which permits us to leave the region of interest, the channel, unperturbed.

Multilevel balancing domain decomposition at extreme scales

María Jesús Samper — Mon, 17 Feb 2020 14:14:37 +0100

In this paper we present a fully distributed, communicator-aware, recursive, and interlevel-overlapped message-passing implementation of the multilevel balancing domain decomposition by constraints (MLBDDC) preconditioner. The implementation highly relies on subcommunicators in order to achieve the desired effect of coarse-grain overlapping of computation and communication, and communication and communication among levels in the hierarchy (namely, interlevel overlapping). Essentially, the main communicator is split into as many nonoverlapping subsets of message-passing interface (MPI) tasks (i.e., MPI subcommunicators) as levels in the hierarchy. Provided that specialized resources (cores and memory) are devoted to each level, a careful rescheduling and mapping of all the computations and communications in the algorithm lets a high degree of overlapping be exploited among levels. All subroutines and associated data structures are expressed recursively, and therefore MLBDDC preconditioners with an arbitrary number of levels can be built while re-using significant and recurrent parts of the codes. This approach leads to excellent weak scalability results as soon as level-1 tasks can fully overlap coarser-levels duties. We provide a model to indicate how to choose the number of levels and coarsening ratios between consecutive levels and determine qualitatively the scalability limits for a given choice. We have carried out a comprehensive weak scalability analysis of the proposed implementation for the three-dimensional Laplacian and linear elasticity problems on structured and unstructured meshes. Excellent weak scalability results have been obtained up to 458,752 IBM BG/Q cores and 1.8 million MPI being, being the first time that exact domain decomposition preconditioners (only based on sparse direct solvers) reach these scales. (An erratum is attached.)

Mixed finite element methods with convection stabilization for the large eddy simulation of incompressible turbulent flows

María Jesús Samper — Mon, 17 Feb 2020 13:40:03 +0100

The variational multiscale method thought as an implicit large eddy simulation model for turbulent flows has been shown to be an alternative to the widely used physical-based models. This method is traditionally combined with equal-order velocity–pressure pairs, since it provides pressure stabilization. In this work, we consider a different approach, based on inf–sup stable elements and convection-only stabilization. In order to do so, we consider a symmetric projection stabilization of the convective term using an orthogonal subscale decomposition. The accuracy and efficiency of this method compared with residual-based algebraic subgrid scales and orthogonal subscales methods for equal-order interpolation is assessed in this paper. Moreover, when inf–sup stable elements are used, the grad–div stabilization term has been shown to be essential to guarantee accurate solutions. Hence, a study of the influence of such term in the large eddy simulation of turbulent incompressible flows is also performed. Furthermore, a recursive block preconditioning strategy has been considered for the resolution of the problem with an implicit treatment of the projection terms. Two different benchmark tests have been solved: the Taylor–Green Vortex flow with Re=1600, and the Turbulent Channel Flow at Reτ=395 and Reτ=590.

On the scalability of inexact balancing domain decomposition by constraints with overlapped coarse/fine corrections

María Jesús Samper — Mon, 17 Feb 2020 13:26:56 +0100

In this work, we analyze the scalability of inexact two-level balancing domain decomposition by constraints (BDDC) preconditioners for Krylov subspace iterative solvers, when using a highly scalable asynchronous parallel implementation where fine and coarse correction computations are overlapped in time. This way, the coarse-grid problem can be fully overlapped by fine-grid computations (which are embarrassingly parallel) in a wide range of cases. Further, we consider inexact solvers to reduce the computational cost/complexity and memory consumption of coarse and local problems and boost the scalability of the solver. Out of our numerical experimentation, we conclude that the BDDC preconditioner is quite insensitive to inexact solvers. In particular, one cycle of algebraic multigrid (AMG) is enough to attain algorithmic scalability. Further, the clear reduction of computing time and memory requirements of inexact solvers compared to sparse direct ones makes possible to scale far beyond state-of-the-art BDDC implementations. Excellent weak scalability results have been obtained with the proposed inexact/overlapped implementation of the two-level BDDC preconditioner, up to 93,312 cores and 20 billion unknowns on JUQUEEN. Further, we have also applied the proposed setting to unstructured meshes and partitions for the pressure Poisson solver in the backward-facing step benchmark domain.

A highly scalable parallel implementation of balancing domain decomposition by constraints

María Jesús Samper — Mon, 17 Feb 2020 12:06:03 +0100

In this work we propose a novel parallelization approach of two-level balancing domain decomposition by constraints preconditioning based on overlapping of fine-grid and coarse-grid duties in time. The global set of MPI tasks is split into those that have fine-grid duties and those that have coarse-grid duties, and the different computations and communications in the algorithm are then re-scheduled and mapped in such a way that the maximum degree of overlapping is achieved while preserving data dependencies among them. In many ranges of interest, the extra cost associated to the coarse-grid problem can be fully masked by fine-grid related computations (which are embarrassingly parallel). Apart from discussing code implementation details, the paper also presents a comprehensive set of numerical experiments, that includes weak scalability analyses, with structured and unstructured meshes, and exact and inexact solvers for the 3D Poisson and linear elasticity problems on a pair of state-of-the-art multicore-based distributed-memory machines. This experimental study reveals remarkable weak scalability in the solution of problems with thousands of millions of unknowns on several tens of thousands of computational cores.

Implementation and Scalability Analysis of Balancing Domain Decomposition Methods

María Jesús Samper — Mon, 17 Feb 2020 11:55:35 +0100

In this paper we present a detailed description of a high-performance distributed-memory implementation of balancing domain decomposition preconditioning techniques. This coverage provides a pool of implementation hints and considerations that can be very useful for scientists that are willing to tackle large-scale distributed-memory machines using these methods. On the other hand, the paper includes a comprehensive performance and scalability study of the resulting codes when they are applied for the solution of the Poisson problem on a large-scale multicore-based distributed-memory machine with up to 4096 cores. Well-known theoretical results guarantee the optimality (algorithmic scalability) of these preconditioning techniques for weak scaling scenarios, as they are able to keep the condition number of the preconditioned operator bounded by a constant with fixed load per core and increasing number of cores. The experimental study presented in the paper complements this mathematical analysis and answers how far can these methods go in the number of cores and the scale of the problem to still be within reasonable ranges of efficiency on current distributed-memory machines. Besides, for those scenarios where poor scalability is expected, the study precisely identifies, quantifies and justifies which are the main sources of inefficiency.

Enhanced balancing Neumann-Neumann preconditioning in computational fluid and solid mechanics

María Jesús Samper — Mon, 17 Feb 2020 11:44:59 +0100

In this work, we propose an enhanced implementation of balancing Neumann-Neumann (BNN) preconditioning together with a detailed numerical comparison against the balancing domain decomposition by constraints (BDDC) preconditioner. As model problems, we consider the Poisson and linear elasticity problems. On one hand, we propose a novel way to deal with singular matrices and pseudo-inverses appearing in local solvers. It is based on a kernel identication strategy that allows us to eciently compute the action of the pseudo-inverse via local indenite solvers. We further show how, identifying a minimum set of degrees of freedom to be xed, an equivalent denite system can be solved instead, even in the elastic case. On the other hand, we propose a simple modication of the preconditioned conjugate gradient (PCG) algorithm that reduces the number of Dirichlet solvers to only one per iteration, leading to similar computational cost as additive methods. After these improvements of the BNN PCG algorithm, we compare its performance against that of the BDDC preconditioners on a pair of large-scale distributed-memory platforms. The enhanced BNN method is a competitive preconditioner for three-dimensional Poisson and elasticity problems, and outperforms the BDDC method in many cases.

A finite element dynamical nonlinear subscale approximation for the low Mach number flow equations

María Jesús Samper — Mon, 17 Feb 2020 11:34:53 +0100

In this work we propose a variational multiscale finite element approximation of thermally coupled low speed flows. The physical model is described by the low Mach number equations, which are obtained as a limit of the compressible Navier–Stokes equations in the small Mach number regime. In contrast to the commonly used Boussinesq approximation, this model permits to take volumetric deformation into account. Although the former is more general than the latter, both systems have similar mathematical structure and their numerical approximation can suffer from the same type of instabilities.

We propose a stabilized finite element approximation based on the variational multiscale method, in which a decomposition of the approximating space into a coarse scale resolvable part and a fine scale subgrid part is performed. Modeling the subscale and taking its effect on the coarse scale problem into account results in a stable formulation. The quality of the final approximation (accuracy, efficiency) depends on the particular model.

The distinctive features of our approach are to consider the subscales as transient and to keep the scale splitting in all the nonlinear terms. The first ingredient permits to obtain an improved time discretization scheme (higher accuracy, better stability, no restrictions on the time step size). The second ingredient permits to prove global conservation properties. It also allows us to approach the problem of dealing with thermal turbulence from a strictly numerical point of view.

Numerical tests show that nonlinear and dynamic subscales give more accurate solutions than classical stabilized methods.

A variational subgrid scale model for transient incompressible flows

María Jesús Samper — Mon, 17 Feb 2020 10:38:44 +0100

We introduce in this paper a variational subgrid scale model for the solution of the incompressible Navier-Stokes equations. With respect to classical multiscale-based stabilisation techniques, we retain the subgrid scale effects in the convective term and integrate the subgrid scale equation in time. The method is applied to the Navier-Stokes equations in an accelerating frame of reference and with Dirichlet (essential), Neumann (natural) and mixed boundary conditions. The concrete objective of the paper is to test a numerical algorithm for solving the non-linear subgrid scale equation and the introduction of the subgrid scale into the grid scale equation. The performance of the technique is demonstrated through the solution of two numerical examples: one to test the tracking of the subgrid scale in the convection term and the other to investigate the effects of considering the subgrid scale transient.

On the modelling of liquid steel processes

María Jesús Samper — Mon, 17 Feb 2020 10:18:21 +0100

An iterative (k-L)-predictor / (epsilon)- corrector algorithm that models turbulent flow was developed in previous publications. In this paper, the 3D finite element turbulent model was used to analyze the liquid steel movement produced by gravity force, inert gas stirring or electromagnetic force stirring.

A new approach to the analysis of vessel residence time distribution curves

María Jesús Samper — Mon, 17 Feb 2020 10:08:21 +0100

Mathematical models for the evaluation of residence time distribution (RTD) curves on a large variety of vessels are presented. These models have been constructed by combination of different tanks or volumes. In order to obtain a good representation of RTD curves, a new volume (called convection diffusion volume) is introduced. The convection-diffusion volume allows the approximation of different experimental or numerical RTD curves with very simple models. An algorithm has been developed to calculate the parameters of the models for any given set of RTD curve experimental points. Validation of the models is carried out by comparison with experimental RTD curves taken from the literature and with a numerical RTD curve obtained by three-dimensional simulation of the flow inside a tundish.

Applications of a (k–ϵ) model for the analysis of continuous casting processes

María Jesús Samper — Mon, 17 Feb 2020 09:32:23 +0100

The finite element solution of the turbulent Navier–Stokes equations developed via (k–ϵ) turbulence models was addressed in previous publications [1–4], where a (k–L)‐predictor/(ϵ)‐corrector iterative algorithm was developed. It was shown that the developed algorithm is robust and converges for the analyses of different flows without requiring the implementation of ad hoc numerical procedures. The turbulent convection–diffusion transport equations are solved by using the velocity distributions determined from the solution of the turbulent Navier–Stokes equations. The dispersion of a die in a turbulent flow can therefore be modelled and the obtained dispersion patterns are validated via flow visualizations in water models. In the present paper, the developed analysis capability is applied to the analysis of continuous casting processes.

A Generalized Finite Difference-Volume Hybrid Method Applied to Shallow-Water Equations

Gerardo Tinoco-Guerrero — Fri, 14 Feb 2020 18:50:26 +0100

Due to the importance of the shallow-water equations in models of real-life phenomena, in recent years the study and model of problems that involve them have been the object of interest of many people. By reason of this, it is imperative to have efficient numerical methods to obtain an approximation of the solutions of the shallow-water equations.

Several authors have worked in approximations using the well-known finite volume and finite element methods, nevertheless, even when these methods compute good approximations to real-life behavior, the computational cost is usually high, which could be a limitation to the application of these methods.

This paper presents an explicit Generalized Finite Difference-Volume Hybrid approximation to the solution of the shallow-water equations, solved on irregular regions meshed with logically rectangular grids; the numerical results show the accuracy obtained with a low-cost implementation. The proposed scheme is a hybridization of a generalized finite difference scheme with the finite volume method.

Coupled effective stress analysis of insertion problems in geotechnics with the Particle Finite Element Method

María Jesús Samper — Fri, 14 Feb 2020 15:15:15 +0100

This paper describes a computational framework for the numerical analysis of quasi-static soil-structure insertion problems in water saturated media. The Particle Finite Element Method is used to solve the linear momentum and mass balance equations at large strains. Solid-fluid interaction is described by a simplified Biot formulation using pore pressure and skeleton displacements as basic field variables. The robustness and accuracy of the proposal is numerically demonstrated presenting results from two benchmark examples. The first one addresses the consolidation of a circular footing on a poroelastic soil. The second one is a parametric analysis of the cone penetration test (CPTu) in a material described by a Cam-clay hyperelastic model, in which the influence of permeability and contact roughness on test results is assessed.

Hydraulic conductivity from piezocone on-the-fly: a numerical evaluation

María Jesús Samper — Fri, 14 Feb 2020 15:03:18 +0100

Permeability is important in many geotechnical applications. The current (cone penetration test that gathers piezometer data (CPTu)) practice to obtain permeability values relies on dissipation tests, which are frequently slow and only linked to permeability through compressibility measures. On-the-fly methods offer an alternative approach in which permeability is directly linked to CPTu penetration measurements. Several on-the-fly methods have been proposed and their applicability and relative advantages are not fully clear. Numerical effective stress simulation of CPTu testing is used here to explore in a simplified but realistic setting the relative merits of different on-the-fly methods. It is found that for partly drained materials, the original simpler relation between cone metrics and normalised permeability works reasonably well. A continuous generalisation of Elsworth and Lee method to the full permeability range is proposed, noting the connection to the backbone normalised pore-pressure curve that describes the partly drained transition of cone penetration. The importance of an undrained limit beyond which the method produces large errors is emphasised.

Numerical Methods for the Modelling of Chip Formation

María Jesús Samper — Fri, 14 Feb 2020 14:18:11 +0100

The modeling of metal cutting has proved to be particularly complex due to the diversity of physical phenomena involved, including thermo-mechanical coupling, contact/friction and material failure. During the last few decades, there has been significant progress in the development of numerical methods for modeling machining operations. Furthermore, the most relevant techniques have been implemented in the relevant commercial codes creating tools for the engineers working in the design of processes and cutting devices. This paper presents a review on the numerical modeling methods and techniques used for the simulation of machining processes. The main purpose is to identify the strengths and weaknesses of each method and strategy developed up-to-now. Moreover the review covers the classical Finite Element Method covering mesh-less methods, particle-based methods and different possibilities of Eulerian and Lagrangian approaches.

Low‐order stabilized finite element for the full Biot formulation in soil mechanics at finite strain

María Jesús Samper — Fri, 14 Feb 2020 14:01:13 +0100

This article presents a novel finite element formulation for the Biot equation using low‐order elements. Additionally, an extra degree of freedom is introduced to treat the volumetric locking steaming from the effective response of the medium; its balance equation is also stabilized. The accuracy of the proposed formulation is demonstrated by means of numerical analyses.

Reply to the Discussion on “Coupled effective stress analysis of insertion problems in geotechnics with the Particle Finite Element method”

María Jesús Samper — Fri, 14 Feb 2020 13:52:49 +0100

The discussers ask why a hyperelastic model was employed. There are several reasons why using an hyper-elastic model is preferable. Probably the most important one is that closed cycles do not produce nor dissipate energy [17, 8]; instead, using a hypoelastic model, closed cycles may produce or dissipate energy, which is inconsistent with the definition of an elastic deformation path. As stressed by Borja et al.[3], the use of an hypoelastic law with constant Poisson’s ratio in the context of the Modified Cam Clay may produce that closed-loop stress paths may result in a non-conservative elastic behavior; therefore, at the end of the loading path the elastic deformation might not be fully recovered. Also, hyper-elastic laws underpin the formulation of large-strain elastoplasticity [17]. By using this theory, the constitutive model is inherently frame-indiferent [17] and algorithmic objective [9].

A stable mesh-independent approach for numerical modelling of structured soils at large strains

María Jesús Samper — Fri, 14 Feb 2020 13:40:54 +0100

We describe the large strain implementation of an elasto-plastic model for structured soils into G-PFEM, a code developed for geotechnical simulations using the Particle Finite Element Method. The constitutive model is appropriate for naturally structured clays, cement-improved soils and soft rocks. Structure may result in brittle behavior even in contractive paths; as a result, localized failure modes are expected in most applications. To avoid the pathological mesh-dependence that may accompany strain localization, a nonlocal reformulation of the model is employed. The resulting constitutive model is incorporated into a numerical code by means of a local explicit stress integration technique. To ensure computability this is hosted within a more general Implicit-Explicit integration scheme (IMPLEX). The good performance of these techniques is illustrated by means of element tests and boundary value problems.

An implicit surface tension model for the analysis of droplet dynamics

María Jesús Samper — Fri, 14 Feb 2020 12:15:40 +0100

A Lagrangian incompressible fluid flow model is extended by including an implicit surface tension term in order to analyze droplet dynamics. The Lagrangian framework is adopted to model the fluid and track its boundary, and the implicit surface tension term is used to introduce the appropriate forces at the domain boundary. The introduction of the tangent matrix corresponding to the surface tension force term ensures enhanced stability of the derived model. Static, dynamic and sessile droplet examples are simulated to validate the model and evaluate its performance. Numerical results are capable of reproducing the pressure distribution in droplets, and the advancing and receding contact angles evolution for droplets in varying substrates and inclined planes. The model is stable even at time steps up to 20 times larger than previously reported in literature and achieves first and second order convergence in time and space, respectively. The present implicit surface tension implementation is applicable to any model where the interface is represented by a moving boundary mesh.

Computational evaluation of cochlear implant surgery outcomes accounting for uncertainty and parameter variability

María Jesús Samper — Fri, 14 Feb 2020 12:11:34 +0100

Cochlear implantation (CI) is a complex surgical procedure that restores hearing in patients with severe deafness. The successful outcome of the implanted device relies on a group of factors, some of them unpredictable or difficult to control. Uncertainties on the electrode array position and the electrical properties of the bone make it difficult to accurately compute the current propagation delivered by the implant and the resulting neural activation. In this context, we use uncertainty quantification methods to explore how these uncertainties propagate through all the stages of CI computational simulations. To this end, we employ an automatic framework, encompassing from the finite element generation of CI models to the assessment of the neural response induced by the implant stimulation. To estimate the confidence intervals of the simulated neural response, we propose two approaches. First, we encode the variability of the cochlear morphology among the population through a statistical shape model. This allows us to generate a population of virtual patients using Monte Carlo sampling and to assign to each of them a set of parameter values according to a statistical distribution. The framework is implemented and parallelized in a High Throughput Computing environment that enables to maximize the available computing resources. Secondly, we perform a patient-specific study to evaluate the computed neural response to seek the optimal post-implantation stimulus levels. Considering a single cochlear morphology, the uncertainty in tissue electrical resistivity and surgical insertion parameters is propagated using the Probabilistic Collocation method, which reduces the number of samples to evaluate. Results show that bone resistivity has the highest influence on CI outcomes. In conjunction with the variability of the cochlear length, worst outcomes are obtained for small cochleae with high resistivity values. However, the effect of the surgical insertion length on the CI outcomes could not be clearly observed, since its impact may be concealed by the other considered parameters. Whereas the Monte Carlo approach implies a high computational cost, Probabilistic Collocation presents a suitable trade-off between precision and computational time. Results suggest that the proposed framework has a great potential to help in both surgical planning decisions and in the audiological setting process.

Analysis of uncertainty and variability in finite element computational models for biomedical engineering: characterization and propagation

María Jesús Samper — Fri, 14 Feb 2020 12:00:50 +0100

Computational modeling has become a powerful tool in biomedical engineering thanks to its potential to simulate coupled systems. However, real parameters are usually not accurately known, and variability is inherent in living organisms. To cope with this, probabilistic tools, statistical analysis and stochastic approaches have been used. This article aims to review the analysis of uncertainty and variability in the context of finite element modeling in biomedical engineering. Characterization techniques and propagation methods are presented, as well as examples of their applications in biomedical finite element simulations. Uncertainty propagation methods, both non-intrusive and intrusive, are described. Finally, pros and cons of the different approaches and their use in the scientific community are presented. This leads us to identify future directions for research and methodological development of uncertainty modeling in biomedical engineering.

Ram-air parachute simulation with panel methods and staggered coupling

María Jesús Samper — Fri, 14 Feb 2020 11:49:12 +0100

Numerical study of droplet dynamics in a polymer electrolyte fuel cell gas channel using an embedded Eulerian-Lagrangian approach

María Jesús Samper — Fri, 14 Feb 2020 11:45:06 +0100

An embedded Eulerian-Lagrangian formulation for the simulation of droplet dynamics within a polymer electrolyte fuel cell (PEFC) channel is presented. Air is modeled using an Eulerian formulation, whereas water is described with a Lagrangian framework. Using this framework, the gas-liquid interface can be accurately identified. The surface tension force is computed using the curvature defined by the boundary of the Lagrangian mesh. The method naturally accounts for material property changes across the interface and accurately represents the pressure discontinuity. A sessile drop in a horizontal surface, a sessile drop in an inclined plane and droplets in a PEFC channel are solved for as numerical examples and compared to experimental data. Numerical results are in excellent agreement with experimental data. Numerical results are also compared to results obtained with the semi-analytical model previously developed by the authors in order to discuss the limitations of the semi-analytical approach.

On the application of the PFEM to droplet dynamics modeling in fuel cells

María Jesús Samper — Fri, 14 Feb 2020 11:02:37 +0100

The Particle Finite Element Method (PFEM) is used to develop a model to study two-phase flow in fuel cell gas channels. First, the PFEM is used to develop the model of free and sessile droplets. The droplet model is then coupled to an Eulerian, fixed-grid, model for the airflow. The resulting coupled PFEM-Eulerian algorithm is used to study droplet oscillations in an air flow and droplet growth in a low-temperature fuel cell gas channel. Numerical results show good agreement with predicted frequencies of oscillation, contact angle, and deformation of injected droplets in gas channels. The PFEM-based approach provides a novel strategy to study droplet dynamics in fuel cells.

A second-order semi-Lagrangian particle FEM (SL-PFEM) method for the incompressible Navier-Stokes equations

Julio García-Espinosa — Thu, 13 Feb 2020 20:01:02 +0100

The Semi-Lagrangian Particle Finite Element Method (SL-PFEM) is a numerical method tailored for solving the fluid dynamics equations. Despite of its excellent numerical properties, such as a minimum numerical erosion in the convective transport and that it exhibits great stability, it has not yet received much attention in the scientific literature. In this presentation, a second order SL-PFEM scheme for solving the incompressible Navier-Stokes equations is presented. This scheme is based on the second order velocity Verlet algorithm, which uses an explicit integration for the particle’s trajectory and an implicit integration for the velocity. The algorithm is completed with a predictor-multicorrector scheme for the integration of the velocity correction using the Finite Element Method. A second order projector based on least squares is used to transfer the intrinsic variables information from the particles onto the background mesh, while a second order interpolation scheme is used to transfer the pressure gradient and viscous accelerations from the mesh to the particles.

Industrial Application of Genetic Algorithms to Cost Reduction of a Wind Turbine Equipped with a Tuned Mass Damper

María Jesús Samper — Thu, 13 Feb 2020 14:53:06 +0100

Design optimization has already become an important tool in industry. The benefits are clear, but several drawbacks are still present, being the main one the computational cost. The numerical simulation involved in the solution of each evaluation is usually costly, but time and computational resources are limited. Time is key in industry. The present communication focuses on the methodology applied to optimize the installation and design of a Tuned Mass Damper. It is a structural device installed within the tower of a wind turbine aimed to stabilize the oscillations and reduce the tensions and the fatigue loads. The paper describes the decision process to define the optimization problem, as well as the issues and solutions applied to deal with a huge computational cost.

Monte Carlo-Based and Sampling-Based Methods and Their Range of Applicability

María Jesús Samper — Thu, 13 Feb 2020 14:48:35 +0100

The present section will focus on the applicability issues of Monte Carlo-based methods, as well as those methods based on sampling techniques. Special focus will be put on the Multi-Level Monte Carlo method and the two implementations developed during the UMRIDA project, namely the Continuous MLMC and MLMC. All named methods have been described in the above sections of this book.

Summary of UMRIDA Best Practices

María Jesús Samper — Thu, 13 Feb 2020 14:43:24 +0100

Uncertainty quantification (UQ) is becoming a strategic step in the design phase. Robust Design Optimization (RDO) is the following step. The Technological Readiness Level (TRL) of intrusive and non-intrusive methodologies is increasing rapidly, although several limitations remain. Nowadays, UQ is a major trend in research, because there is a lot of room for improvement.

General Introduction to Monte Carlo and Multi-level Monte Carlo Methods

María Jesús Samper — Thu, 13 Feb 2020 14:39:52 +0100

In this chapter, we present a general introduction to Monte Carlo (MC)-based methods, sampling methodologies, stratification methods, and variance reduction techniques. In the first part, we will discuss the theoretical basis and the convergence proprieties of MC methods. The next part is devoted to pseudorandom and quasi-random number generation, the generation of random variables and the application of stratification. It is followed by techniques for correlation and discrepancy control. The third part presents the concept of Latin Hypercube Sampling (LHS). The last part introduces the concept of Multi-Level Monte Carlo (MLMC).

Multi-level Monte Carlo Method

María Jesús Samper — Thu, 13 Feb 2020 14:33:21 +0100

Uncertainty quantification has gained interest during the recent years. Two clear examples are NODESIM-CFD and, the just finished, UMRIDA projects.

Industrial application of genetic algorithms to cost reduction of a wind turbine equipped with a tuned mass damper

María Jesús Samper — Thu, 13 Feb 2020 14:23:15 +0100

Description of the Test Cases

María Jesús Samper — Thu, 13 Feb 2020 14:11:50 +0100

The high-level objective of MARS project was to understand the formation and behaviour of turbulent structures which affects the Reynolds stress and skin friction. The aim was, once understood, to apply flow control techniques in order to control these structures and reduce the overall drag derived from the Reynolds stress and mainly from the skin friction. Active flow control devices were the main interest; DBD plasma, Synthetic jets, Micro Blowing and Suction, Moving Surfaces were included on the list. To test all these devices, two test cases were defined, and a database and file repository were established in the project webserver. The present chapter is aimed to describe the test cases, including the set-up of the flow control devices, as well as to describe the file repository were all the data was stored.

Optimization of the Experimental Set-up for a Turbulent Separated Shear Flow Control by Plasma Actuator Using Genetic Algorithms

María Jesús Samper — Thu, 13 Feb 2020 14:00:27 +0100

Since 1947, when Schubauer and Skramstad established the basis of the technology with its revolutionary work about steady state tools and mechanisms for the flow management, the progress of the flow control technology and the development of devices have progressed constantly. Anyway, the applicability of such devices is limited, and only few of them have arrived to the assembly workshop. The problem is that the range of actuation is still limited. Despite their operability limitations, flow control devices are of great interest for the aeronautical industry. The number of projects investigating this technology demonstrates the relevance of in the Fluid Dynamic field. The scientific interest focus not only on the industrial applications and the improvement of the technology, but also on the deep understanding of the physical phenomena associated to the flow separation, turbulence formation associated to the final drag reduction aim. A clear example of what has been mentioned is the EC MARS research project (MARS project, FP7 project number 266326, [2]). Its objectives are aimed to a better understanding of the Reynolds Stress and turbulent flow related to both drag reduction and flow control. The research was carried out through the analysis of several flow control devices and the optimization of the parameters for some of them was an important element of the research. When solving a traditional fluid dynamics optimisation problem numerical flow analysis are used instead of experimental ones due to their lower cost and shorter needed time for evaluation of candidate solutions. Nevertheless, in the particular case of the selected flow control plasma devices the experimental measurement of the performance of each candidate configuration has been much quicker than a numerical analysis. For this reason, the corresponding optimisation problem has been solved by coupling an evolutionary optimization algorithm with an experimental device. This paper discusses the design quality and efficiency gained by this innovative coupling.

Lessons learned associated with constructability application in a massive housing project

Luisa Casadei — Thu, 13 Feb 2020 11:05:03 +0100

The architecture, engineering and construction industry has great activity and importance, it moves other areas to satisfy the infrastructure needs of most economic and social activities, such as housing, roads, education, industry and health, among others However, it is one of the industries with the lowest productivity and which seems not to have taken advantage, as do other production sectors in general, of the opportunities offered by some management philosophies to solve complications and difficulties. This article analyzes the most relevant lessons learned from the implementation, both in the previous phases and in the link or relations between them, and in the corporate area of the company, when applying the constructablity in the planning stages, design and acquisitions, guiding the objectives of multidisciplinary teams towards the integration of a housing project built in Barquisimeto, Venezuela. The methodology was used in the integral process of the project, prior to the preliminary project, which led to the continuation of benefits in costs, times and quality of the final product in a pilot project.

Drag reduction via turbulent boundary layer flow control

María Jesús Samper — Thu, 13 Feb 2020 10:35:52 +0100

Turbulent boundary layer control (TBLC) for skin-friction drag reduction is a relatively new technology made possible through the advances in computational-simulation capabilities, which have improved the understanding of the flow structures of turbulence. Advances in micro-electronic technology have enabled the fabrication of active device systems able to manipulating these structures. The combination of simulation, understanding and micro-actuation technologies offers new opportunities to significantly decrease drag, and by doing so, to increase fuel efficiency of future aircraft. The literature review that follows shows that the application of active control turbulent skin-friction drag reduction is considered of prime importance by industry, even though it is still at a low technology readiness level (TRL). This review presents the state of the art of different technologies oriented to the active and passive control for turbulent skin-friction drag reduction and contributes to the improvement of these technologies.

Turbulent separated shear flow control by surface plasma actuator: experimental optimization by genetic algorithm approach

María Jesús Samper — Thu, 13 Feb 2020 10:18:17 +0100

The potential benefits of active flow control are no more debated. Among many others applications, flow control provides an effective mean for manipulating turbulent separated flows. Here, a nonthermal surface plasma discharge (dielectric barrier discharge) is installed at the step corner of a backward-facing step (U 0 = 15 m/s, Re h = 30,000, Re θ = 1650). Wall pressure sensors are used to estimate the reattaching location downstream of the step (objective function #1) and also to measure the wall pressure fluctuation coefficients (objective function #2). An autonomous multi-variable optimization by genetic algorithm is implemented in an experiment for optimizing simultaneously the voltage amplitude, the burst frequency and the duty cycle of the high-voltage signal producing the surface plasma discharge. The single-objective optimization problems concern alternatively the minimization of the objective function #1 and the maximization of the objective function #2. The present paper demonstrates that when coupled with the plasma actuator and the wall pressure sensors, the genetic algorithm can find the optimum forcing conditions in only a few generations. At the end of the iterative search process, the minimum reattaching position is achieved by forcing the flow at the shear layer mode where a large spreading rate is obtained by increasing the periodicity of the vortex street and by enhancing the vortex pairing process. The objective function #2 is maximized for an actuation at half the shear layer mode. In this specific forcing mode, time-resolved PIV shows that the vortex pairing is reduced and that the strong fluctuations of the wall pressure coefficients result from the periodic passages of flow structures whose size corresponds to the height of the step model.

Review: Course in structural engineering. Regional center for technological developments for construction, seismology and seismic engineering CeDeReTec

Luisa Casadei — Thu, 13 Feb 2020 10:00:03 +0100

This is the review of the Structural Engineering Update course carried out at the Regional Center for Technological Developments for Seismology and Seismic Engineering Construction (CeDeReTec, Mendoza, Argentina) at the end of the year 2019. The presentations are shown and also reports on the activities carried out within the framework of the Project “Destructive Potential Generated by the Earthquake of Ecuador of April 16, 2016” are reported.