Related papers: An Efficient Surrogate Model of Secondary Electron…
The problem of self-consistently coupling kinetic runaway-electron physics to the macroscopic evolution of the plasma is addressed by dividing the electron population into a bulk and a tail. A probabilistic closure is adopted to determine…
This doctoral thesis covers some of my advances in electron microscopy with deep learning. Highlights include a comprehensive review of deep learning in electron microscopy; large new electron microscopy datasets for machine learning,…
This paper presents an enhanced version of the subgroup method for resonance self-shielding treatment, termed the robust subgroup method, which integrates Robust Estimation (RE) with a Differential Evolution (DE) algorithm. The RE approach…
In many mechanistic medical, biological, physical and engineered spatiotemporal dynamic models the numerical solution of partial differential equations (PDEs) can make simulations impractically slow. Biological models require the…
A deep-learning-based surrogate model capable of predicting flow and geomechanical responses in CO2 storage operations is presented and applied. The 3D recurrent R-U-Net model combines deep convolutional and recurrent neural networks to…
A surrogate model approximates the outputs of a solver of Partial Differential Equations (PDEs) with a low computational cost. In this article, we propose a method to build learning-based surrogates in the context of parameterized PDEs,…
We present the first successful simulation of a induced disruption in ASDEX Upgrade from massive material injection (MMI) up to established runaway electron (RE) beam, thus covering pre-thermal quench, thermal quench and current quench (CQ)…
Evolutionary Reinforcement Learning (ERL) that applying Evolutionary Algorithms (EAs) to optimize the weight parameters of Deep Neural Network (DNN) based policies has been widely regarded as an alternative to traditional reinforcement…
Modeling the time-dependent evolution of electron density is essential for understanding quantum mechanical behaviors of condensed matter and enabling predictive simulations in spectroscopy, photochemistry, and ultrafast science. Yet, while…
The disruption and runaway electron analysis model code was extended to include tungsten impurities in disruption simulations with the aim of studying the runaway electron (RE) generation. This study investigates RE current sensitivity on…
Runaway electrons (RE) generated during magnetic disruptions present a major threat to the safe operation of plasma nuclear fusion reactors. A critical aspect of understanding RE dynamics is to calculate the runaway probability, i.e., the…
Machine learning provides a novel avenue for the study of experimental realizations of many-body systems, and has recently been proven successful in analyzing properties of experimental data of ultracold quantum gases. We here show that…
Microstructural evolution is a key aspect of understanding and exploiting the structure-property-performance relation of materials. Modeling microstructure evolution usually relies on coarse-grained simulations with evolution principles…
The development of a reliable and robust surrogate model is often constrained by the dimensionality of the problem. For a system with high-dimensional inputs/outputs (I/O), conventional approaches usually use a low-dimensional manifold to…
Plasma current instabilities can destabilize the plasma discharge and cool the plasma rapidly. In such $\textit{disruptions}$ or in the start-up phase of the reactor, inductive electric fields are generated which accelerate electrons to…
Laws of electrodynamics constrain scattering cross-sections of resonant objects. Nevertheless, a fundamental bound that expresses how larger that scattering cross-section can be is yet to be found. Approaches based on cascading multiple…
We present predictions for the distribution of rapidity gaps in realistic kinematics of future electron-ion colliders, based on numerical solutions of the original Kovchegov-Levin equation and of its next-to-leading extension taking into…
Collision avoidance algorithms are essential for safe and efficient robot operation among pedestrians. This work proposes using deep reinforcement (RL) learning as a framework to model the complex interactions and cooperation with nearby,…
We present an improved version of the Superscaling Analysis with Relativistic Effective Mass, denoted as SuSAM-v2. In the original SuSAM model, a universal scaling function was fitted to a selected set of quasielastic electron scattering…
Surrogate models for partial-differential equations are widely used in the design of meta-materials to rapidly evaluate the behavior of composable components. However, the training cost of accurate surrogates by machine learning can rapidly…