A meshless quasicontinuum method based on local maximum-entropy interpolation

被引:19
|
作者
Kochmann, Dennis M. [1 ]
Venturini, Gabriela N. [1 ]
机构
[1] CALTECH, Grad Aerosp Labs, Pasadena, CA 91125 USA
关键词
molecular dynamics; crystalline solids; quasicontinuum; meshfree methods; MESHFREE APPROXIMATION SCHEMES; DISLOCATION NUCLEATION; MOLECULAR-DYNAMICS; BRIDGING SCALE; LENGTH SCALES; SIMULATION; NANOINDENTATION; MECHANICS; MODELS; HOMOGENIZATION;
D O I
10.1088/0965-0393/22/3/034007
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Coarse-graining atomistic ensembles can overcome the practical limitations of molecular statics and dynamics in order to facilitate simulations at much larger length scales than accessible by discrete atomistic techniques due to computational expense. The quasicontinuum (QC) method was introduced to reduce the number of degrees of freedom in crystalline solids by choosing a set of representative atoms from the fully atomistic ensemble and obtaining the positions and momenta of all remaining lattice sites by interpolation. Here, we present a new energy-based nonlocal meshless version of the QC method based on local maximum-entropy (max-ent) interpolation schemes instead of the traditional polynomial interpolation, which particularly promises advantages in model adaptation to tie atomistic resolution to crystal defects while efficiently coarse-graining away from these. To this end, we formulate the meshless QC representation and analyze its performance. One-dimensional chain problems allow for clean mathematical treatment and provide interesting insight, which allow us to quantify the approximation error as a function of representative atom distribution and support of meshless shape functions. A fully three-dimensional implementation then demonstrates the applicability of the new QC scheme and highlights its features. Overall, we show that local max-ent interpolation offers a number of advantages over previous QC realizations.
引用
收藏
页数:28
相关论文
共 50 条
  • [1] Adaptive meshless local maximum-entropy finite element method for convection–diffusion problems
    C. T. Wu
    D. L. Young
    H. K. Hong
    Computational Mechanics, 2014, 53 : 189 - 200
  • [2] A stable Maximum-Entropy Meshless method for analysis of porous media
    Nazem, M.
    Kardani, M.
    Bienen, B.
    Cassidy, M.
    COMPUTERS AND GEOTECHNICS, 2016, 80 : 248 - 260
  • [3] A MAXIMUM-ENTROPY APPROACH TO INTERPOLATION
    FATHIMA, ST
    YEGNANARAYANA, B
    SIGNAL PROCESSING, 1990, 21 (01) : 17 - 24
  • [4] Adaptive meshless local maximum-entropy finite element method for convection-diffusion problems
    Wu, C. T.
    Young, D. L.
    Hong, H. K.
    COMPUTATIONAL MECHANICS, 2014, 53 (01) : 189 - 200
  • [5] Adaptive meshless local maximum-entropy finite element method for Navier-Stokes equations
    Young, D. L.
    Shih, C-L
    Yen, L. J.
    Chu, C-R
    JOURNAL OF MECHANICS, 2024, 40 : 475 - 490
  • [6] THE MAXIMUM-ENTROPY METHOD
    SELIGMAN, TH
    LECTURE NOTES IN PHYSICS, 1985, 236 : 78 - 91
  • [7] THE MAXIMUM-ENTROPY METHOD
    SMITH, DJ
    MARCONI REVIEW, 1981, 44 (222): : 137 - 158
  • [8] THE MAXIMUM DETERMINANT METHOD AND THE MAXIMUM-ENTROPY METHOD
    NARAYAN, R
    NITYANANDA, R
    ACTA CRYSTALLOGRAPHICA SECTION A, 1982, 38 (JAN): : 122 - 128
  • [9] THE MAXIMUM-ENTROPY METHOD IN NMR
    MARTIN, JF
    JOURNAL OF MAGNETIC RESONANCE, 1985, 65 (02) : 291 - 297
  • [10] The maximum-entropy method in superspace
    van Smaalen, S
    Palatinus, L
    Schneider, M
    ACTA CRYSTALLOGRAPHICA A-FOUNDATION AND ADVANCES, 2003, 59 : 459 - 469