Neural-network potential energy surface with small database and high precision: A benchmark of the H + H2 system

2019 ◽  
Vol 151 (11) ◽  
pp. 114302 ◽  
Author(s):  
Qingfei Song ◽  
Qiuyu Zhang ◽  
Qingyong Meng
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
High Precision ◽  
Energy Surface

An accurate full-dimensional potential energy surface for the reaction OH + SO → H + SO2

2021 ◽  
Vol 23 (1) ◽  
pp. 487-497
Author(s):  
Jie Qin ◽  
Jun Li
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Energy Surface ◽  
Invariant Polynomial ◽  
Network Approach ◽  
Neural Network Approach ◽  
Polynomial Neural Network

An accurate full-dimensional PES for the OH + SO ↔ H + SO2 reaction is developed by the permutation invariant polynomial-neural network approach.

scholarly journals Using principal component analysis for neural network high-dimensional potential energy surface

2020 ◽  
Vol 152 (23) ◽  
pp. 234103
Author(s):  
Bastien Casier ◽  
Stéphane Carniato ◽  
Tsveta Miteva ◽  
Nathalie Capron ◽  
Nicolas Sisourat
Keyword(s):  
Neural Network ◽  
Principal Component Analysis ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Energy Surface ◽  
Principal Component ◽  
Component Analysis ◽  
High Dimensional

An accurate full-dimensional permutationally invariant potential energy surface for the interaction between H2O and CO

2019 ◽  
Vol 21 (43) ◽  
pp. 24101-24111 ◽  
Author(s):  
Yang Liu ◽  
Jun Li
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Energy Surface ◽  
Invariant Polynomial ◽  
Polynomial Neural Network ◽  
Invariant Potential

The first full-dimensional accurate potential energy surface was developed for the CO + H2O system based on ca. 102 000 points calculated at the CCSD(T)-F12a/AVTZ level using a permutation invariant polynomial-neural network (PIP-NN) method.

A neural network potential energy surface for the NaH2 system and dynamics studies on the H(2S) + NaH(X1Σ+) → Na(2S) + H2(X1Σg+) reaction

2017 ◽  
Vol 19 (30) ◽  
pp. 19873-19880 ◽  
Author(s):  
Shufen Wang ◽  
Jiuchuang Yuan ◽  
Huixing Li ◽  
Maodu Chen
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Energy Surface ◽  
Neural Network Method ◽  
The Neural Network ◽  
Network Method ◽  
High Level

A new potential energy surface of the NaH2 system is obtained using the neural network method based on high-level energies.

A new potential energy surface for the ground electronic state of the LiH2 system, and dynamics studies on the H(2S) + LiH(X1Σ+) → Li(2S) + H2(X1Σg+) reaction

2015 ◽  
Vol 17 (17) ◽  
pp. 11732-11739 ◽  
Author(s):  
Jiuchuang Yuan ◽  
Di He ◽  
Maodu Chen
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Electronic State ◽  
Energy Surface ◽  
Ground Electronic State ◽  
Neural Network Method ◽  
Network Method ◽  
High Level

A new potential energy surface of the LiH2 system is obtained using a neural network method based on high-level energies.

scholarly journals Automated Construction of Neural Network Potential Energy Surface: The Enhanced Self-Organizing Incremental Neural Network Deep Potential Method

2021 ◽  
Author(s):  
Mingyuan Xu ◽  
Tong Zhu ◽  
John ZH Zhang
Keyword(s):  
Neural Network ◽  
Potential Energy ◽  
Potential Energy Surface ◽  
Potential Energy Surfaces ◽  
Energy Surface ◽  
Chemical Space ◽  
Potential Method ◽  
Dynamics Simulation ◽  
Data Set ◽  
Self Organizing

In recent years, the use of deep learning (neural network) potential energy surface (NNPES) in molecular dynamics simulation has experienced explosive growth as it can be as accurate as quantum chemistry methods while being as efficient as classical mechanic methods. However, the development of NNPES is highly non-trivial. In particular, it has been troubling to construct a dataset that is as small as possible yet can cover the target chemical space. In this work, an ESOINN-DP method is developed, which has the enhanced self-organizing incremental neural-network (ESOINN) and a newly proposed error indicator at its core. With ESOINN-DP, One can construct the NNPES with little human intervention, and this method ensures that the constructed reference dataset covers the target chemical space with minimum redundancy. The performance of the ESOINN-DP method has been well validated by developing neural network potential energy surfaces for water clusters and by de-redundancy of a sub-data set of the ANI-1 database. We believe that the ESOINN-DP method provides a novelty idea for the construction of NNPES and especially, the reference datasets, and it can be used for MD simulations of various gas-phase and condensed-phase chemical systems.

Sign in / Sign up

Export Citation Format

Share Document