Mixed strong form representation particle method for solids and structures

David Tae, Kumar K Tamma

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, a generalized particle system (GPS) method, a general method to describe multiple strong form representation based particle methods is described. Gradient, divergence, and Laplacian operators used in various strong form based particle method such as moving particle semi-implicit (MPS) method, smooth particle hydrodynamics (SPH), and peridynamics, can be described by the GPS method with proper selection of parameters. In addition, the application of mixed formulation representation to the GPS method is described. Based on Hu- Washizu principle and Hellinger-Reissner principle, the mixed form refers to the method solving multiple primary variables such as displacement, strain and stress, simultaneously in the FEM method; however for convenience in employing FEM with particle methods, a simple representation in construction only is shown. It is usually applied to finite element method (FEM) to overcome numerical errors including locking issues. While the locking issues do not arise in strong form based particle methods, the mixed form representation in construction only concept applied to GPS method can be the first step for fostering coupling of multi-domain problems, coupling mixed form FEM and mixed form representation GPS method; however it is to be noted that the standard GPS particle method and the mixed for representation construction GPS particle method are equivalent. Two dimensional simple bar and beam problems are presented and the results from mixed form GPS method is comparable to the mixed form FEM results.

Original languageEnglish (US)
Pages (from-to)429-441
Number of pages13
JournalJournal of Applied and Computational Mechanics
Volume4
Issue numberSpecialissue
DOIs
StatePublished - Sep 1 2018

Fingerprint

Finite element method
Hydrodynamics

Keywords

  • Mixed form
  • Particle method
  • Structural mechanics/dynamics

Cite this

Mixed strong form representation particle method for solids and structures. / Tae, David; Tamma, Kumar K.

In: Journal of Applied and Computational Mechanics, Vol. 4, No. Specialissue, 01.09.2018, p. 429-441.

Research output: Contribution to journalArticle

@article{9ed4f14ff4644e00ae7bfe5e5ac11790,
title = "Mixed strong form representation particle method for solids and structures",
abstract = "In this paper, a generalized particle system (GPS) method, a general method to describe multiple strong form representation based particle methods is described. Gradient, divergence, and Laplacian operators used in various strong form based particle method such as moving particle semi-implicit (MPS) method, smooth particle hydrodynamics (SPH), and peridynamics, can be described by the GPS method with proper selection of parameters. In addition, the application of mixed formulation representation to the GPS method is described. Based on Hu- Washizu principle and Hellinger-Reissner principle, the mixed form refers to the method solving multiple primary variables such as displacement, strain and stress, simultaneously in the FEM method; however for convenience in employing FEM with particle methods, a simple representation in construction only is shown. It is usually applied to finite element method (FEM) to overcome numerical errors including locking issues. While the locking issues do not arise in strong form based particle methods, the mixed form representation in construction only concept applied to GPS method can be the first step for fostering coupling of multi-domain problems, coupling mixed form FEM and mixed form representation GPS method; however it is to be noted that the standard GPS particle method and the mixed for representation construction GPS particle method are equivalent. Two dimensional simple bar and beam problems are presented and the results from mixed form GPS method is comparable to the mixed form FEM results.",
keywords = "Mixed form, Particle method, Structural mechanics/dynamics",
author = "David Tae and Tamma, {Kumar K}",
year = "2018",
month = "9",
day = "1",
doi = "10.22055/jacm.2018.24877.1216",
language = "English (US)",
volume = "4",
pages = "429--441",
journal = "Journal of Applied and Computational Mechanics",
issn = "2383-4536",
publisher = "Shahid Chamran University of Ahvaz",
number = "Specialissue",

}

TY - JOUR

T1 - Mixed strong form representation particle method for solids and structures

AU - Tae, David

AU - Tamma, Kumar K

PY - 2018/9/1

Y1 - 2018/9/1

N2 - In this paper, a generalized particle system (GPS) method, a general method to describe multiple strong form representation based particle methods is described. Gradient, divergence, and Laplacian operators used in various strong form based particle method such as moving particle semi-implicit (MPS) method, smooth particle hydrodynamics (SPH), and peridynamics, can be described by the GPS method with proper selection of parameters. In addition, the application of mixed formulation representation to the GPS method is described. Based on Hu- Washizu principle and Hellinger-Reissner principle, the mixed form refers to the method solving multiple primary variables such as displacement, strain and stress, simultaneously in the FEM method; however for convenience in employing FEM with particle methods, a simple representation in construction only is shown. It is usually applied to finite element method (FEM) to overcome numerical errors including locking issues. While the locking issues do not arise in strong form based particle methods, the mixed form representation in construction only concept applied to GPS method can be the first step for fostering coupling of multi-domain problems, coupling mixed form FEM and mixed form representation GPS method; however it is to be noted that the standard GPS particle method and the mixed for representation construction GPS particle method are equivalent. Two dimensional simple bar and beam problems are presented and the results from mixed form GPS method is comparable to the mixed form FEM results.

AB - In this paper, a generalized particle system (GPS) method, a general method to describe multiple strong form representation based particle methods is described. Gradient, divergence, and Laplacian operators used in various strong form based particle method such as moving particle semi-implicit (MPS) method, smooth particle hydrodynamics (SPH), and peridynamics, can be described by the GPS method with proper selection of parameters. In addition, the application of mixed formulation representation to the GPS method is described. Based on Hu- Washizu principle and Hellinger-Reissner principle, the mixed form refers to the method solving multiple primary variables such as displacement, strain and stress, simultaneously in the FEM method; however for convenience in employing FEM with particle methods, a simple representation in construction only is shown. It is usually applied to finite element method (FEM) to overcome numerical errors including locking issues. While the locking issues do not arise in strong form based particle methods, the mixed form representation in construction only concept applied to GPS method can be the first step for fostering coupling of multi-domain problems, coupling mixed form FEM and mixed form representation GPS method; however it is to be noted that the standard GPS particle method and the mixed for representation construction GPS particle method are equivalent. Two dimensional simple bar and beam problems are presented and the results from mixed form GPS method is comparable to the mixed form FEM results.

KW - Mixed form

KW - Particle method

KW - Structural mechanics/dynamics

UR - http://www.scopus.com/inward/record.url?scp=85057856534&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057856534&partnerID=8YFLogxK

U2 - 10.22055/jacm.2018.24877.1216

DO - 10.22055/jacm.2018.24877.1216

M3 - Article

AN - SCOPUS:85057856534

VL - 4

SP - 429

EP - 441

JO - Journal of Applied and Computational Mechanics

JF - Journal of Applied and Computational Mechanics

SN - 2383-4536

IS - Specialissue

ER -