Abstract:Multiscale Modeling &Simulation, Volume 22, Issue 4, Page 1478-1497, December 2024. We employ intelligent dissipative particle dynamics that combines dissipative particle dynamics and machine learning techniques to perform the coarse graining of a water molecule system, including single- and multiple-water coarse graining. This coarse graining is a challenging task due to the high-density nature of water, the presence of long-range electrostatic interaction, and the difficulty of packing multiple molecules. A TIP3P model is used to simulate water in molecular dynamics simulations, providing the raw data for the machine learning model. The machine learning model predicts the conservative and dissipative force fields used in dissipative particle dynamics to replicate the static and dynamic properties in the molecular dynamics simulation. A constrained K-means clustering method is used to dynamically pack an equal number of water molecules into a coarse graining particle, ensuring uniform size and mass for every particle. The six coarse graining levels are examined to determine whether both the static and dynamic properties are replicated for the single-water and multiple-water models. The simulation results show that the static properties of water, such as temperature, radial distribution function, and velocity probability density function, can be successfully preserved by the present coarse graining method at each level. The dynamic properties, such as viscosity and diffusion, are difficult to preserve due to an overestimated friction force, and depend on a tunable parameter called time correlation limit, such that these properties can be successfully replicated by choosing a proper time correlation limit for each coarse graining level.

Machine learning coarse-grained models of dissolutive wetting: a droplet on soluble surfaces

Coarse Graining of Multiple Water Molecules with Intelligent Dissipative Particle Dynamics

Machine learning assisted coarse-grained molecular dynamics modeling of meso-scale interfacial fluids

Machine learning potentials for complex aqueous systems made simple

Machine-assisted quantification of droplet boiling upon multiple solid materials

Multiscale Simulation of Dynamic Wetting

Neural Network-Assisted Model of Interfacial Fluids with Explicit Coarse-Grained Molecular Structures.

A Machine Learning Model of Liquid Bridge Force and Its Application in Discrete Element Method

Droplets on Micro Structured or Patterned Surfaces with Different Roughness Parameters

A Machine Learning Approach to Quantify Dissolution Kinetics of Porous Media

Molecular-Dynamics Simulations of Droplets on A Solid Surface

Molecular dynamics simulation of wetting behavior at CO2/water/solid interfaces

A multiscale volume of fluid method with self-consistent boundary conditions derived from molecular dynamics

Revisiting the Application of Machine Learning Approaches in Predicting Aqueous Solubility

Molecular dynamics simulations of the evaporation of particle-laden droplets

Multiple Wetting–Dewetting States of a Water Droplet on Dual-Scale Hierarchical Structured Surfaces

An electro-hydrodynamics modeling of droplet actuation on solid surface by surfactant-mediated electro-dewetting

Wetting Behaviours of a Single Droplet on Biomimetic Micro Structured Surfaces

Prediction of Water Transport Properties on an Anisotropic Wetting Surface Via Deep Learning.

Application of Molecular Simulation on Wettability and Its Alteration

Machine Learning of coarse-grained Molecular Dynamics Force Fields