Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 30 for “"Machine learning potentials"”.
-
Modelling chemical processes with machine learning potentials
Machine learning potentials (MLPs) are increasingly being considered as an alternative to density functional theory (DFT) for driving molecular dynamics (MD) simulations due to their comparable accuracy and lower computational cost. However, their application in studying chemical processes, …
-
Developments of Machine Learning Potentials for Atomistic Simulations
… cases. In this thesis, I will show the power of machine learning potentials, resolving the predicaments described above. First, the machine learning potential methods will be applied to SiO2 for investigating the implications of different machine learning potentials. Second, the machine learning …
-
Machine-learning potentials for finitetemperature simulations of polarons
… work, polarons are modelled in MgO and LiF using machine-learned interatomic potentials (MLIPs) to enable finite-temperature calculations at high speed. At 0 K, LiF exhibits two types of stable polarons. The transition between the two polaron states is first studied with hybrid density functional …
-
Simulating chemical reactivity using machine learning potentials and umbrella sampling
… exploring bond breaking or forming processes. Machine learning potentials (MLPs) have emerged as a promising tool for exploring dynamics at the accuracy of quantum mechanics and a fraction of the cost of AIMD. However, training often requires many data points and expert knowledge. In this …
-
Dispersion interactions in machine learning potentials for large-scale atomistic simulations
… atomistic simulations. With the recent advent of machine learning (ML) in materials science, the problem with the computational cost of the DFT calculations can be mitigated by training ML potentials for various materials using DFT data as the reference database. These potentials are able to …
-
Understanding the Behavior of Machine Learning Potentials for Simulation of Gas Adsorption
… to maintain both accuracy and transferability. Machine learning (ML) potential offers a promising middle ground, combining near-QM accuracy with significantly improved computational performance. In this study, we systematically explore how accurate ML potential should be, when it is used in …
-
The Structural Information Filtered Features Potential for Machine Learning calculations of energies and forces of atomic systems.
In the last ten years, machine learning potentials have been successfully applied to the study of crystals, and molecules. However, more complex materials like clusters, macro-molecules, and glasses are out reach of current methods. The input of any machine learning system is a tensor of features …
-
Combining Artificial Intelligence With Physics Based Simulations in Understanding and Design of Interfaces
… simulations with advanced techniques like machine learning (ML) and data-driven approaches is crucial. This dissertation investigates the challenges and opportunities associated with utilizing simulation and ML techniques to study interfacial phenomena and design chemically patterned …
-
Probing Ion Effects In Nanoconfined Aqueous Electrolytes: A Molecular Dynamics Study Using Neural Network Potentials
… To address these challenges, we employ a machine learning approach to develop a many-body potential that simultaneously achieves ab initio accuracy and high throughput in molecular dynamics simulations. In this work, we compare the structure and dynamics of aqueous NaCl and KCl salts when …
-
Development and Assessment of Interatomic Neural Network Potentials for Reactive Chemistry and Molecular Dynamics Simulations
… and computational cost in their calculations. Machine learning methods are on the rise to solve this trade off by training on large datasets of highly accurate calculations that are traditionally hard to obtain. The development of interatomic machine learning potentials has resulted in the …
-
Machine learning-aided chemical kinetic modeling
… scale of systems and the need for interatomic potentials that ensure first-principle accuracy in describing molecular-level processes. The evolution of machine learning techniques has expanded the boundaries of system and time scales in theoretical chemistry, a breakthrough previously …
-
Characterising the energy landscapes of molecular organic crystals
… the energy landscapes of organic crystals using machine learning potentials are detailed. These involve active learning from low-level predicted landscapes as well as on-the-fly training within Monte Carlo simulations. Potentials trained with these methods on a variety of compounds are shown to …
-
Exploring active surfaces of ZnxCryOz for syngas conversion via MLPs simulation
… true active surfaces. Recently, the emergence of machine learning potentials (MLPs) has offered a solution by combining the computational efficiency of empirical methods with the accuracy of density functional theory (DFT) calculations. This advancement enables the exploration of vast structural …
-
Particle velocities and deposition for micro-cold spray and cold spray processes
… Parallel Simulator). A variety of interatomic potentials are used to simulate the behavior of the Ta/Ta₂O₅ material system including embedded atom method (EAM), variable charge potentials, and custom machine learning potentials using the Deep Potential for Molecular Dynamics kit (DeePMD-Kit). …
-
REACTIVE FORCE FIELD BASED ATOMISTIC SIMULATIONS OF SILICON ANODE UPON LITHIATION AND DELITHIATION IN LITHIUM-ION BATTERIES
… extensive training set of Li-Si system for machine-learning potentials development. There are two major discoveries based on the simulation work. First, to elucidate the rate-limiting factor upon lithiation of Si for improved rate performance, reactive MD simulations were performed in …
-
Data-driven analysis and atomistic simulations of next-generation materials for energy conversion and storage
… theory (DFT) computations are combined with machine learning (ML) regressors and classifiers for ranking and property prediction tasks. By employing carefully selected descriptors, these models achieve formation energy predictions with 95% accuracy and reliably distinguish between metallic …
Page 1 of 2