Fast plane wave density functional theory molecular dynamics calculations on multi-GPU machines

Jia, Weile; Fu, Jiyun; Cao, Zongyan; Wang, Long; Chi, Xuebin; Gao, Weiguo; Wang, Lin‐Wang

doi:10.1016/j.jcp.2013.05.005

Cited by 322 publications

(157 citation statements)

References 18 publications

Supporting

Mentioning

146

Contrasting

Unclassified

Order By: Relevance

“…Compared to the traditional HDNNP method using only the total energy of the system, SANNP acquires hundreds of times more energy information from the same DFT. As a result, we found that only 1000 MD steps (which takes about half day to finish using GPU by the PWmat code [35][36][37]) is sufficient to train an accurate SANNP, which dramatically reduces the training time while dramatically increasing the accuracy.…”

Section: Resultsmentioning

confidence: 99%

Density functional theory based neural network force fields from energy decompositions

et al. 2019

Self Cite

View full text Add to dashboard Cite

In order to develop force fields (FF) for molecular dynamics simulations that retain the accuracy of ab initio density functional theory (DFT), we developed a machine learning (ML) protocol based on an energy decomposition scheme that extracts atomic energies from DFT calculations.Our DFT to FF (DFT2FF) approach provides almost hundreds of times more data for the DFT energies, which dramatically improves accuracy with less DFT calculations. In addition, we use piecewise cosine basis functions to systematically construct symmetry invariant features into the neural network model. We illustrate this DFT2FF approach for amorphous silicon where only 800 DFT configurations are sufficient to achieve an accuracy of 1meV/atom for energy and 0.1eV/A for forces. We then use the resulting FF model to calculate the thermal conductivity of amorphous Si based on long molecular dynamics simulations. The dramatic speedup in training 2 in our DFT2FF protocol allows the adoption of a new simulation paradigm where an accurate and problem specific FF for a given physics phenomenon is trained on-the-spot through a quick DFT pre-calculation and FF training.

show abstract

Section: Resultsmentioning

confidence: 99%

Density functional theory based neural network force fields from energy decompositions

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…The DFT calculations were performed by using PWmat [25][26][27]. NCPP-SG15-PBE pseudopotential [28,29] was used with a 50 Ryd cutoff in plane wave basis set.…”

Section: àmentioning

confidence: 99%

Discovering unusual structures from exception using big data and machine learning techniques

Jie

Qian

et al. 2019

Science Bulletin

Self Cite

View full text Add to dashboard Cite

a b s t r a c tRecently, machine learning (ML) has become a widely used technique in materials science study. Most work focuses on predicting the rule and overall trend by building a machine learning model. However, new insights are often learnt from exceptions against the overall trend. In this work, we demonstrate that how unusual structures are discovered from exceptions when machine learning is used to get the relationship between atomic and electronic structures based on big data from high-throughput calculation database. For example, after training an ML model for the relationship between atomic and electronic structures of crystals, we find AgO 2 F, an unusual structure with both Ag 3+ and O 2 2À , from structures whose band gap deviates much from the prediction made by our model. A further investigation on this structure might shed light into the research on anionic redox in transition metal oxides of Li-ion batteries.

show abstract

“…Finally, band structures and density of states (DOS) calculations were performed for these defective models using PWmat soware. 25,26 3. Results Fig.…”

Section: Computational Studiesmentioning

confidence: 99%

The aggression behavior study of Cl⁻ on the defect structure of passive films on copper

Dong

Wei

et al. 2019

RSC Adv.

View full text Add to dashboard Cite

show abstract

Fast plane wave density functional theory molecular dynamics calculations on multi-GPU machines

Cited by 322 publications

References 18 publications

Density functional theory based neural network force fields from energy decompositions

Density functional theory based neural network force fields from energy decompositions

Discovering unusual structures from exception using big data and machine learning techniques

The aggression behavior study of Cl⁻ on the defect structure of passive films on copper

Contact Info

Product

Resources

About

Fast plane wave density functional theory molecular dynamics calculations on multi-GPU machines

Cited by 322 publications

References 18 publications

Density functional theory based neural network force fields from energy decompositions

Density functional theory based neural network force fields from energy decompositions

Discovering unusual structures from exception using big data and machine learning techniques

The aggression behavior study of Cl− on the defect structure of passive films on copper

Contact Info

Product

Resources

About

The aggression behavior study of Cl⁻ on the defect structure of passive films on copper