A5079469780- Molten salt chemistry and electrochemical processes
- Machine Learning in Materials Science
- Thermal Expansion and Ionic Conductivity
- Phase Change Materials Research
- Advanced Thermoelectric Materials and Devices
- Computational Drug Discovery Methods
- Material Dynamics and Properties
- Advanced Battery Materials and Technologies
- Nuclear Physics and Applications
- Nuclear Materials and Properties
- Protein Structure and Dynamics
- Thermal properties of materials
- ZnO doping and properties
- Chemical Looping and Thermochemical Processes
- Advancements in Solid Oxide Fuel Cells
- Nuclear reactor physics and engineering
- Gas Sensing Nanomaterials and Sensors
- Adsorption and Cooling Systems
- Inorganic Fluorides and Related Compounds
- Hydrogen Storage and Materials
- Nuclear physics research studies
- Astro and Planetary Science
- Metallurgical Processes and Thermodynamics
- Zeolite Catalysis and Synthesis
- Ga2O3 and related materials
East China University of Science and Technology
2019-2022
Chongqing Normal University
2011
DeePMD-kit is a powerful open-source software package that facilitates molecular dynamics simulations using machine learning potentials known as Deep Potential (DP) models. This package, which was released in 2017, has been widely used the fields of physics, chemistry, biology, and material science for studying atomistic systems. The current version offers numerous advanced features, such DeepPot-SE, attention-based hybrid descriptors, ability to fit tensile properties, type embedding, model...
Theoretical studies on the MgCl2–KCl eutectic heavily rely ab initio calculations based density functional theory (DFT). However, neither large-scale nor long-time are feasible in framework of method, which makes it challenging to accurately predict some properties. To address this issue, a scheme calculation, deep neural networks, and machine learning is introduced. By training high-quality data sets generated by calculations, potential (DP) constructed describe interaction between atoms....
Abstract In previous work, molten magnesium chloride has been investigated using first‐principles molecular dynamics (FPMD) simulations based on density functional theory (DFT). However, such are computationally intensive and therefore restricted in terms of simulated size time. this a machine learning‐based deep potential (DP) is trained to accelerate the simulation chloride. The DP can accurately describe energies forces with prediction errors energy force being 1.76 × 10 −3 eV/atom 4.76...
We perform deep variational free energy calculations to investigate the dense hydrogen system at 1200 K and high pressures. In this computational framework, neural networks are used model through proton Boltzmann distribution electron wavefunction. By directly minimizing energy, our results reveal emergence of a crystalline order associated with center mass molecules approximately 180 GPa. This transition from atomic liquid molecular solid is marked by discontinuities in both pressure...
Abstract The marriage of ab initio calculations and machine learning (ML) methods exhibits bright application prospects in interatomic potential development. In this work, a concurrent scheme is implemented to automatically generate ML for the MgCl 2 ‐NaCl eutectic. This allows train with training datasets approximately four times smaller than that previous which significantly reduces computational cost. learned used accelerate estimation properties eutectic, thermal conductivity particular....
DeePMD-kit is a powerful open-source software package that facilitates molecular dynamics simulations using machine learning potentials (MLP) known as Deep Potential (DP) models. This package, which was released in 2017, has been widely used the fields of physics, chemistry, biology, and material science for studying atomistic systems. The current version offers numerous advanced features such DeepPot-SE, attention-based hybrid descriptors, ability to fit tensile properties, type embedding,...