A5067761186- Advanced Numerical Methods in Computational Mathematics
- Matrix Theory and Algorithms
- Model Reduction and Neural Networks
- Electromagnetic Scattering and Analysis
- Computational Fluid Dynamics and Aerodynamics
- Plasma Diagnostics and Applications
- Plasma Applications and Diagnostics
- Advanced Optimization Algorithms Research
- Numerical methods for differential equations
- Lattice Boltzmann Simulation Studies
- Electromagnetic Simulation and Numerical Methods
- Numerical methods in engineering
- Iterative Methods for Nonlinear Equations
- Advanced Mathematical Modeling in Engineering
- Electrohydrodynamics and Fluid Dynamics
- Particle accelerators and beam dynamics
- Ionosphere and magnetosphere dynamics
- Electrostatic Discharge in Electronics
- Lightning and Electromagnetic Phenomena
- Meteorological Phenomena and Simulations
- Microfluidic and Capillary Electrophoresis Applications
- Genetics, Bioinformatics, and Biomedical Research
- Tensor decomposition and applications
- Advanced MRI Techniques and Applications
- Composite Material Mechanics
National Central University
2015-2024
University of Macau
2022
Nanjing University of Science and Technology
2022
King Abdullah University of Science and Technology
2022
Korea Institute of Industrial Technology
2014
University of Colorado Boulder
2004-2006
University of Colorado System
2004
University of Colorado Denver
2002
Abstract Ionospheric total electron content (TEC) is a key indicator of the space environment. Geophysical forcing from above and below drives its spatial temporal variations. A full understanding physical chemical principles, available well‐representable driving inputs, capable computational power are required for models to reproduce simulations that agree with observations, which may be challenging at times. Recently, data‐driven approaches, such as deep learning, have therefore surged...
The full-space Lagrange--Newton algorithm is one of the numerical algorithms for solving problems arising from optimization constrained by nonlinear partial differential equations. Newton-type methods enjoy fast convergence when nonlinearity in system well-balanced; however, some problems, such as control incompressible flows, even linear difficult to achieve and a long stagnation period often appears iteration history. In this work, we introduce nonlinearly preconditioned inexact Newton...
Tongue diagnosis, a noninvasive examination, is an essential step for syndrome differentiation and treatment in traditional Chinese medicine (TCM). Sublingual vein (SV) examined to determine the presence of blood stasis syndrome. Many studies have shown that degree SV positively correlates with disease severity. However, diagnoses examination are often subjective because they influenced by factors such as physicians' experience color perception, resulting different interpretations....
Abstract We introduce a new submesh strategy for the two‐level finite element method. The numerical results show that is able to better capture boundary layer which caused by choice of bubble functions. effect an improved approximation residual free bubbles studied advective–diffusive equation. Copyright © 2002 John Wiley & Sons, Ltd.
Currency exchange rate prediction is a typical time series problem which has been solved by time-series models, such as Autoregressive Integrated Moving Average (ARIMA), Seasonal ARIMA (SARIMA) well machine learning methods, Single Layer Perception (SLP) and Long Short-Term Memory (LSTM). In this work, we aim to predict the future currency prices in collaboration with news sentiment analysis. We use Australian dollar (AUD) against US case study AUD for next day, week, two-week, month....
A parallel 2D axisymmetric plasma fluid modeling for an inductively coupled source with tetrafluoromethane precursor is reported. In total, 32 species 96 gas‐phase and 27 surface reactions site‐balance equations are considered. The predicted results of major densities in reasonable agreement reported experiments. etching products, e.g. SiF x O 2 , found to be appreciable (∼10%) compared the near substrate. density trends, such as CF + ( = 1–3), also consistent Finally, rate on SiO substrate...
A nonlinear elimination preconditioned inexact Newton (NEPIN) algorithm is proposed for problems with localized strong nonlinearities. Due to unbalanced nonlinearities (``nonlinear stiffness''), the traditional method often exhibits a long plateau in norm of residual or even fails converge. NEPIN implicitly removes components causing trouble global convergence through correction based on within subspace that provides modified direction iteration. Numerical experiments show can be more robust...
To analyze the saturation performance of IEEE 802.11 DCF, several discrete-time Markov chain models for a station and sets channel equations have been introduced. We take into account frame retry limit, freezing backoff counter, dependence procedure on previous status all together. Our method is simple even though it accurate under assumption no consecutive transmissions over shared channel.