An interface shell element for coupling non-matching quadrilateral shell meshes

Ho-Nguyen-Tan, Thuan,Kim, Hyun-Gyu Elsevier 2018 Computers & structures Vol.208 No.-

<P><B>Abstract</B></P> <P>In this study, a novel interface shell element (ISE) is developed based on a variable-node element formulation to couple non-matching quadrilateral shell meshes. Shape functions for ISEs are explicitly presented in a polynomial form with the use of appropriate supports of weight functions in moving least square (MLS) approximation. Assumed natural strains in the form of the mixed interpolation of tensorial components (MITC) approach are employed to avoid the transverse shear locking when the thickness of shell tends to zero. Moreover, an assumed membrane strain field defined over quadrilateral subdomains subdividing an ISE is used to alleviate the membrane locking in curved ISEs. Numerical experiments show the effectiveness and efficiency of ISE for connecting dissimilar quadrilateral shell meshes at a common interface.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Interface shell elements can be easily defined to couple non-matching quadrilateral shell meshes. </LI> <LI> Shape functions for interface shell elements are explicitly presented in a polynomial form. </LI> <LI> The MITC approach is successfully employed to avoid the transverse shear locking of interface shell elements. </LI> <LI> An assumed membrane strain field is used to alleviate the membrane locking in curved interface shell elements. </LI> <LI> Interface shell elements can successfully alleviate locking phenomena and show good convergence for non-matching mesh problems. </LI> </UL> </P>

3차원 유한요소 모델의 불일치 격자 결합을 위한 계면요소

김현규(Hyun-Gyu Kim) 대한기계학회 2007 대한기계학회 춘추학술대회 Vol.2007 No.10

Gluing of a non-matching interface between dissimilar meshes is one of challenging tasks in computational mechanics. Interface element method (IEM) has been developed by using an appropriate interpolation scheme on interface regions between partitioned finite element meshes. The continuity of displacements at the interfaces is satisfied exactly, and the completeness of the shape functions for the interface elements provides a reasonable transfer of strain fields through the non-matching interfaces. Moreover, the compatibility of the interface elements and the finite elements provides a seamless connection of dissimilar meshes. The construction of the interface elements makes it possible to connect dissimilar meshes with arbitrarily curved non-matching interfaces in three dimensions.

유체-고체 상호작용 문제를 위한 미끌림 격자 알고리즘

김현규(Hyun-Gyu Kim) 대한기계학회 2009 대한기계학회 춘추학술대회 Vol.2009 No.5

The study concerns a sliding mesh algorithm to the problem of the computation of fluid-solid interactions with moving boundaries. The interface elements(IE) are considered to couple non-conforming meshes between fluid and solid finite elements. Conditions of compatibility and continuity between fluid and solid meshes are satisfied exactly by constructing interface meshes in accordance with a sliding interface. An arbitrary Lagrangian-Eulerian(ALE) description is adopted for the fluid domain, while for the solid domain an updated Lagrangian formulation is considered to accommodate finite deformations of an elastic structure. Numerical tests are proposed for fluid-solid interaction problems involving a sling non-matching interface.

Application of the Runge Kutta Discontinuous Galerkin-Direct Ghost Fluid Method to internal explosion inside a water-filled tube

Jinwon Park 대한조선학회 2019 International Journal of Naval Architecture and Oc Vol.11 No.1

This paper aims to assess the applicability of the Runge Kutta Discontinuous Galerkin-Direct Ghost Fluid Method to the internal explosion inside a water-filled tube, which previously was studied by many researchers in separate works. Once the explosive charge located at the inner center of the water-filled tube explodes, the tube wall is subjected to an extremely high intensity fluid loading and deformed. The deformation causes a modification of the field of fluid flow in the region near the water-structure interface so that has substantial influence on the response of the structure. To connect the structure and the fluid, valid data exchanges along the interface are essential. Classical fluid structure interaction simulations usually employ a matched meshing scheme which discretizes the fluid and structure domains using a single mesh density. The computational cost of fluid structure interaction simulations is usually governed by the structure because the size of time step may be determined by the density of structure mesh. The finer mesh density, the better solution, but more expensive computational cost. To reduce such computational cost, a non-matched meshing scheme which allows for different mesh densities is employed. The coupled numerical approach of this paper has fewer difficulties in the implementation and computation, compared to gas dynamics based approach which requires complicated analytical manipulations. It can also be applied to wider compressible, inviscid fluid flow analyses often found in underwater explosion events.

Application of the Runge Kutta Discontinuous Galerkin-Direct Ghost Fluid Method to internal explosion inside a water-filled tube

Park, Jinwon The Society of Naval Architects of Korea 2019 International Journal of Naval Architecture and Oc Vol.11 No.1

This paper aims to assess the applicability of the Runge Kutta Discontinuous Galerkin-Direct Ghost Fluid Method to the internal explosion inside a water-filled tube, which previously was studied by many researchers in separate works. Once the explosive charge located at the inner center of the water-filled tube explodes, the tube wall is subjected to an extremely high intensity fluid loading and deformed. The deformation causes a modification of the field of fluid flow in the region near the water-structure interface so that has substantial influence on the response of the structure. To connect the structure and the fluid, valid data exchanges along the interface are essential. Classical fluid structure interaction simulations usually employ a matched meshing scheme which discretizes the fluid and structure domains using a single mesh density. The computational cost of fluid structure interaction simulations is usually governed by the structure because the size of time step may be determined by the density of structure mesh. The finer mesh density, the better solution, but more expensive computational cost. To reduce such computational cost, a non-matched meshing scheme which allows for different mesh densities is employed. The coupled numerical approach of this paper has fewer difficulties in the implementation and computation, compared to gas dynamics based approach which requires complicated analytical manipulations. It can also be applied to wider compressible, inviscid fluid flow analyses often found in underwater explosion events.

Development of polygonal shell element and its application to coupling non-matching meshes

Thuan Ho-Nguyen-Tan(호응웬탄뚜안),Hyun-Gyu Kim(김현규) 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.5

유체-고체 상호작용 해석을 위한 계면요소의 개발

김현규(Hyun-Gyu Kim) 대한기계학회 2008 대한기계학회 춘추학술대회 Vol.2008 No.11

This paper presents a new approach to simulate fluid-solid interaction problems involving non-matching interfaces. The coupling between fluid and solid domains with dissimilar finite element meshes consisting of 4-node quadrilateral elements is achieved by using the interface element method (IEM). Conditions of compatibility between fluid and solid meshes are satisfied exactly by introducing the interface elements defined on interfacing regions. Importantly, a consistent transfer of loads through matching interface element meshes guarantees the present method to be an efficient approach of the solution strategy to fluid-solid interaction problems. An arbitrary Lagrangian-Eulerian (ALE) description is adopted for the fluid domain, while for the solid domain an updated Lagrangian formulation is considered to accommodate finite deformations of an elastic structure. The stabilized equal order velocity-pressure elements for incompressible flows are used in the motion of fluids. Fully coupled equations are solved simultaneously in a single computational domain. Numerical results are presented for fluid-solid interaction problems involving nonmatching interfaces to demonstrate the effectiveness of the methodology.

MLS-based variable-node elements compatible with quadratic interpolation. Part I: formulation and application for non-matching meshes

Cho, Young-Sam,Im, Seyoung John Wiley Sons, Ltd. 2006 International journal for numerical methods in eng Vol.65 No.4

<P>Two-dimensional variable-node elements compatible with quadratic interpolation are developed using the moving least-square (MLS) approximation. The mapping from the parental domain to the physical element domain is implicitly obtained from MLS approximation, with the shape functions and their derivatives calculated and saved only at the numerical integration points. It is shown that the present MLS-based variable-node elements meet the patch test if a sufficiently large number of integration points are employed for numerical integration. The cantilever problem with non-matching meshes is chosen to check the feasibility of the present MLS-based variable-node elements, and the result is compared with that from the lower-order case compatible with linear interpolation. Copyright © 2005 John Wiley & Sons, Ltd.</P>

Development of an interface shell element for coupling non-matching meshes

Thuan Ho-Nguyen-Tan(호응웬탄뚜안),Hyun-Gyu Kim(김현규) 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.11

Mortar formulation for a class of staggered discontinuous Galerkin methods

Kim, H.H.,Chung, E.T.,Lam, C.Y. Pergamon Press ; Elsevier Science Ltd 2016 COMPUTERS & MATHEMATICS WITH APPLICATIONS - Vol.71 No.8

<P>A mortar formulation is developed and analyzed for a class of staggered discontinuous Galerkin (SDG) methods applied to second order elliptic problems in two dimensions. The computational domain consists of nonoverlapping subdomains and a triangulation is provided for each subdomain, which need not conform across subdomain interfaces. This feature allows a more flexible design of discrete models for problems with complicated geometries, shocks, or singular points. A mortar matching condition is enforced on the solutions across the subdomain interfaces by introducing a Lagrange multiplier space. Moreover, optimal convergence rates in both L-2 and discrete energy norms are proved. Numerical results are presented to show the performance of the method. (C) 2016 Elsevier Ltd. All rights reserved.</P>