A5023452561- HVDC Systems and Fault Protection
- Energy Load and Power Forecasting
- Geophysical Methods and Applications
- Microwave Imaging and Scattering Analysis
- Microwave Engineering and Waveguides
- Machine Learning and ELM
- Grey System Theory Applications
- Superconducting Materials and Applications
- High-Voltage Power Transmission Systems
- Antenna Design and Analysis
- Power Systems and Renewable Energy
- Antenna Design and Optimization
- Electric Power System Optimization
- Frequency Control in Power Systems
- Electromagnetic Simulation and Numerical Methods
- Physics of Superconductivity and Magnetism
- Advanced Antenna and Metasurface Technologies
- Wind Turbine Control Systems
- Microgrid Control and Optimization
- Electromagnetic Scattering and Analysis
- Ultrasonics and Acoustic Wave Propagation
- Solar Thermal and Photovoltaic Systems
- Power System Optimization and Stability
- Metamaterials and Metasurfaces Applications
- solar cell performance optimization
Xiamen University
2019-2025
Kunming University of Science and Technology
2023-2024
Chinese Academy of Sciences
2008-2024
Institute of Electrical Engineering
2006-2024
University of Chinese Academy of Sciences
2022-2024
Semiconductor Manufacturing International (China)
2024
Tan Kah Kee Innovation Laboratory
2024
University of Electronic Science and Technology of China
2014-2020
National Laboratory for Superconductivity
2016
Chinese Academy of Engineering
2005
In this communication, a novel artificial neural network (ANN) model is proposed to describe the antenna performance with various parameters. model, three parallel and independent branches are involved for different Meanwhile, data-classification technique of support vector machine also included classify geometrical variables into proper categories. Once input, ANN can simultaneously obtain S-parameter, gain, radiation pattern from branches. The validity efficiency confirmed Fabry-Perot...
A low-voltage ride-through (LVRT) strategy for a doubly fed induction generator (DFIG) with switch-type fault current limiter (STFCL) is presented in this paper. The STFCL composed of fault-current-limiting inductors, isolation transformers, diode bridge, semiconductor switch, snubber capacitor, and energy absorption bypass. can insert inductors series the stator branches on occurrence grid fault, which limit rotor overcurrent weaken back electromagnetic force voltage simultaneously. safety...
To improve the convenience and efficiency of antenna design, in this article, a novel inverse artificial neural network (ANN) model is proposed which performance indexes are set as input corresponding geometrical variables output. solve multiobjective problem modeling, first part ANN involves three parallel independent branches for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</i> -parameter, gain, radiation pattern, second outputs final...
Frequency deviation of power systems caused by grid-connected wind fluctuations is one the key factors limiting level penetration. Applying an energy storage system (ESS) to overcome constraint frequency can increase permitted This paper proposes a method evaluate penetration as limited grid and size ESS expected level. The proposed treats stochastic process rather than deterministic signal takes response characteristics into full consideration. Application based on theoretical derivations...
A dynamic adjustment kernel extreme learning machine with transfer functions is proposed for parametric modeling of the electromagnetic behavior microwave components. If satisfactory accuracy has not been obtained, model, which supports functionalities increased learning, reduced and hybrid can utilize overlap between old training data set new one to achieve accurate trained results faster retraining. The validity model confirmed two examples a microstrip-to-microstrip vertical transition...
As one of the most important technologies for next generation very-large scale integrated circuit fabrication, extreme ultraviolet (EUV) lithography has attracted more and attention in recent years.However, EUV lithography, optical distortion printed image on wafer always negative impacts imaging performance.Thus, to enhance performance system, especially small critical dimensions, this work, a novel proximity correction (OPC) system based deep learning technique is proposed.It includes...
To efficiently and conveniently realize the design of frequency selective surface (FSS) structures with many degrees freedoms (DoFs), a spatial inverse method (SIDM) based on machine learning technology is proposed. The proposed SIDM takes advantages modeling topological to spatially for FSS. Different from simple parametric or modeling, which only involves one type variable, i.e. binary continuous variables, contains both variables flexibly model FSS less cost. Meanwhile, multilayer...
A dual-module machine learning scheme is proposed to reconstruct inhomogeneous scatterers with high contrasts and large electrical dimensions. The first nonlinear mapping module (NMM) an extreme (ELM), which used convert the measured scattered fields at receiver arrays into preliminary images of scatterers. second image-enhancing (IEM) a convolutional neural network (CNN), refine further from NMM obtain high-accuracy pixel-based model parameter distribution in inversion domain. Compared...
To solve the restriction of prior knowledge in artificial neural networks (ANNs) for modeling finite periodic arrays, a new multigrade ANN model is proposed this paper. Considering mutual coupling and array environment, designed with two sub-ANNs, element-ANN array-ANN. Based on relationship between geometrical variables electromagnetic (EM) behavior elements an array, built to provide Then, review array-ANN modeled obtain EM response whole from nonlinear superposition element responses....