A Tungsten Deep Neural-Network Potential for Simulating Mechanical Property Degradation Under Fusion Service Environment

Abstract

Tungsten is a promising candidate material in fusion energy facilities. Molecular dynamics (MD) simulations reveal the atomistic scale mechanisms, so they are crucial for the understanding of the macroscopic property deterioration of tungsten under harsh and complex service environments. The interatomic potential used in the MD simulations is required to accurately describe a wide spectrum of relevant defect properties, which is by far challenging to the existing interatomic potentials. In this paper, we propose a new three-body embedding descriptor and hybridize it into the Deep-Potential (DP) framework, an end-to-end deep learning interatomic potential model. Trained with the dataset generated by a concurrent learning method, the potential model for tungsten, named by DP-HYB, is able to accurately predict a wide range of properties including elastic constants, stacking fault energy, the formation energies of free surfaces and point defects, which are included in the training dataset, and formation energies of grain boundaries and prismatic loops, the core structure of screw dislocation, the Peierls barrier and the transition path of the screw dislocation migration, which are not explicitly included in the training dataset. The DP-HYB is a good candidate for the atomistic simulations of tungsten property deterioration, especially those involving the mechanical property degradation under the harsh fusion service environment.

References in corpus (21)

• Adam: A Method for Stochastic Optimization
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models
• Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
• SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
• Deep Potential: a general representation of a many-body potential energy surface
• Machine-learning interatomic potential for radiation damage and defects in tungsten
• The influence of transition metal solutes on dislocation core structure and values of Peierls stress and barrier in tungsten
• Gaussian approximation potentials for body-centered-cubic transition metals
• Deep learning inter-atomic potential model for accurate irradiation damage simulations
• Total Teleportation of an Entangled State
• Assessment of interatomic potentials for atomistic analysis of static and dynamic properties of screw dislocations in W
• Accurate Deep Potential model for the Al-Cu-Mg alloy in the full concentration space
• A modified W-W interatomic potential based on ab initio calculations
• Inflation and the Measurement Problem
• Tungsten migration energy barriers for surface diffusion: a parameterization for KMC simulations

Cited by in corpus (5)

• DeePMD-kit v2: A software package for Deep Potential models
• Unraveling the Crystallization Kinetics of the Ge${}_2$Sb${}_2$Te${}_5$ Phase Change Compound with a Machine-Learned Interatomic Potential
• Classical and Machine Learning Interatomic Potentials for BCC Vanadium
• Exciting DeePMD: Learning excited state energies, forces, and non-adiabatic couplings
• Atomistic understanding of hydrogen bubble-induced embrittlement in tungsten enabled by machine learning molecular dynamics