TY - GEN
T1 - Fast local pressure estimation for two dimensional systems from molecular dynamics simulations
AU - Yesudasan, Sumith
AU - Chacko, Sibi
PY - 2018
Y1 - 2018
N2 - Coupling of statistical properties from atomistic simulationsto continuum is essential to model many multi-scale phenomena.Often, the system under consideration will be homogeneous intwo-dimensions (2-D). But due to the existing coupling methods, the property estimation takes place in three-dimensions (3-D) and then averaged to 2-D, which is computationally expensivedue to the 3-D convolutions. A direct 2-D pressure or stress estimation model is lacking in literature. In this work, we develop adirect 2-D pressure field estimation method which is much fasterthan 3-D methods without losing accuracy. The method is validated with MD simulations on two systems: a liquid film and acylindrical drop of argon suspended in surrounding vapor. Thisformulation will enable the study of 2-D fundamental phenomena like passive liquid flows in microlayer, as well as facilitatethe coupling of atomistic and continuum simulations with reduced computational cost.
AB - Coupling of statistical properties from atomistic simulationsto continuum is essential to model many multi-scale phenomena.Often, the system under consideration will be homogeneous intwo-dimensions (2-D). But due to the existing coupling methods, the property estimation takes place in three-dimensions (3-D) and then averaged to 2-D, which is computationally expensivedue to the 3-D convolutions. A direct 2-D pressure or stress estimation model is lacking in literature. In this work, we develop adirect 2-D pressure field estimation method which is much fasterthan 3-D methods without losing accuracy. The method is validated with MD simulations on two systems: a liquid film and acylindrical drop of argon suspended in surrounding vapor. Thisformulation will enable the study of 2-D fundamental phenomena like passive liquid flows in microlayer, as well as facilitatethe coupling of atomistic and continuum simulations with reduced computational cost.
UR - http://www.scopus.com/inward/record.url?scp=85055452520&partnerID=8YFLogxK
U2 - 10.1115/POWER2018-7263
DO - 10.1115/POWER2018-7263
M3 - Conference contribution
AN - SCOPUS:85055452520
SN - 9780791851401
BT - ASME 2018 Power Conference
PB - American Society of Mechanical Engineers
ER -