A5088849506- Magnetic confinement fusion research
- Laser-Plasma Interactions and Diagnostics
- Fusion materials and technologies
- Ionosphere and magnetosphere dynamics
- Gene expression and cancer classification
- Natural Language Processing Techniques
- Complex Network Analysis Techniques
- Anomaly Detection Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Advanced Graph Neural Networks
- Solar and Space Plasma Dynamics
- Industrial Vision Systems and Defect Detection
- Text Readability and Simplification
- Single-cell and spatial transcriptomics
- Graph Theory and Algorithms
- Cell Image Analysis Techniques
Shanghai Institute for Science of Science
2023
Universidad Carlos III de Madrid
2015-2017
University of South Australia
2010
Abstract Non-Maxwellian distributions of particles are commonly observed in fusion studies, especially for magnetic confinement plasmas. The particle distribution has a direct effect on reactivity, which is the focus this study. We investigate effects three types non-Maxwellian distributions, namely drift-ring-beam, slowing-down, and kappa super-thermal reactivities D-T (Deuterium-Trillium) p-B11 (proton-Boron) using newly developed program, where enhancement reactivity relative to...
The impact of isotope mass on the radial correlation length () and long-range (LRC) plasma turbulence has been investigated in electron cyclotron resonance heated (ECRH) low-density deuterium-hydrogen plasmas TJ-II stellarator. We find that increases with deuterium/hydrogen (D/H) ratio. However, amplitude LRC decreases slightly D/H ratio, contrast to previous results tokamak Xu et al (2013 Phys. Rev. Lett. 110 265005). These findings show effect both largest scales (LRC determined by size...
The effects of 3D geometry are explored in TJ-II from two relevant points view: neoclassical transport and modification stability dispersion relation waves. Particle fuelling impurity studied considering the properties, paying attention to both other possible mechanisms. magnetic topology on stability, confinement Alfvén Eigenmodes properties also explored, showing possibility controlling modes by modifying configuration; onset similar geodesic acoustic driven fast electrons or ions; weak...
Abstract The isotope effect, namely the dependence of plasma confinement, is still one principal scientific conundrums facing magnetic fusion community. We have investigated impact mass on multi-scale mechanisms, including characterization radial correlation lengths ( <?CDATA $\boldsymbol{L}_{{r}}$ ?> <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mstyle displaystyle="false"> <mml:mrow> <mml:msub> <mml:mi mathvariant="bold-italic">L</mml:mi> </mml:mrow>...
Large Language Models (LLMs) have achieved impressive results across numerous NLP tasks but still encounter difficulties in machine translation. Traditional methods to improve translation typically involved fine-tuning LLMs using parallel corpora. However, vanilla often leads catastrophic forgetting of the instruction-following capabilities and alignment with human preferences, compromising their broad general abilities introducing potential security risks. These abilities, which are...
Different biological labs conduct similar experiments on same diseases. It is highly desirable to have a better model based more experimental results than that single result. In this paper, we propose method for integrating microarray data from multiple sources building classification models. To test the method, use three real world sets different with devices and environments. Although well known its inconsistencies across labs, demonstrate it possible build consistent models using labs. We...
Non-Maxwellian distributions of particles are commonly observed in fusion studies, especially for magnetic confinement plasmas. The particle distribution has a direct effect on reactivity, which is the focus this study. We investigate effects three types non-Maxwellian distributions, namely drift-ring-beam, slowing-down, and kappa super-thermal reactivities D-T (Deuterium-Trillium) p-B11 (proton-Boron) using newly developed program, where enhancement reactivity relative to Maxwellian...
Link prediction is a key task in graph data, which used to determine whether there connection between two nodes. Since neural networks (GNNs) can process data efficiently, most link methods are based on GNNs. However, because GNNs require large amount of computing resources and time, they not effective enough for large-scale industrial production. GNN-to-MLP distillation effectively solve this problem, uses as the teacher model train high-performance student MLP. existing mainly focus...