A5036276196- Numerical methods in engineering
- Model Reduction and Neural Networks
- Electromagnetic Scattering and Analysis
- Numerical methods in inverse problems
- Electromagnetic Simulation and Numerical Methods
- Fractional Differential Equations Solutions
- Composite Material Mechanics
- Quantum Chromodynamics and Particle Interactions
- Composite Structure Analysis and Optimization
- Nanofluid Flow and Heat Transfer
- Approximation Theory and Sequence Spaces
- Differential Equations and Numerical Methods
- Advanced Numerical Methods in Computational Mathematics
- Mathematical Approximation and Integration
- Matrix Theory and Algorithms
- Advanced Algorithms and Applications
- Image and Signal Denoising Methods
- Advanced Sensor and Control Systems
- Particle physics theoretical and experimental studies
- Geoscience and Mining Technology
- Cold Atom Physics and Bose-Einstein Condensates
- Geotechnical Engineering and Underground Structures
- Magnetic Properties and Applications
- Mathematical functions and polynomials
- Structural Health Monitoring Techniques
Qingdao University
2022-2025
Northeastern University
2014-2024
Institute of Mechanics
2024
Wuhan University of Technology
2023
Hohai University
2018-2021
Australian National University
2020-2021
State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering
2019
China University of Mining and Technology
2016
North China University of Water Resources and Electric Power
2015-2016
Shanghai Jiao Tong University
2016
In this paper, we aim to numerically resolve linear and nonlinear transient heat conduction problems in multilayer composite materials using a deep learning method called multi-domain physics-informed neural networks (MDPINNs). For purpose, the media are first divided into independent sub-domains based on domain decomposition technique. The single-layer established, each sub-domain has its corresponding sub-network. Then, two sub-networks connected by continuity conditions interface form...
This paper proposed a new physics-informed neural network (PINN) for solving the Hausdorff derivative Poisson equations (HDPEs) on irregular domains by using concept of fractal derivative. The present scheme transforms numerical solution partial differential equation into an optimization problem including governing and boundary conditions. Like meshless method, developed PINN does not require grid generation integration. Moreover, it can freely address non-uniformly distributed nodes. study...
Abstract In this paper, the standard and localized space‐time radial basis function (RBF) collocation methods are modified combined with time‐marching scheme domain decomposition technique for simulating long‐time transient heat conduction in 3D anisotropic composite materials. proposed approaches, we set source points outside whole RBF method or established subdomain approach by introducing space time magnification factors, instead of distributing them inside original domain. addition,...
ABSTRACT This study proposes a hybrid collocation approach for simulating heat conduction problems in anisotropic functionally graded materials over extended time intervals. In this approach, the Krylov deferred correction (KDC) scheme is employed temporal discretization of dynamic problems, featuring novel numerical implementation designed to ensure precise satisfaction boundary conditions. The localized radial basis function (LRBF) method modified and utilized solve resulting value...
The sound barrier is an important means to reduce noise caused by traveling vehicles on roads or railways. Structural design and optimization of the can effectively use materials improve reduction effect. In this paper, a new isogeometric singular boundary method proposed applied shape barriers. geometric structure accurately represented using non-uniform rational B-splines. acoustic sensitivity control points was calculated direct differentiation adjoint variable method. After that, moving...
Image sharpening based on the partial differential equations plays an important role in fields of image processing. It is effective technique to clear and sharpen features, provides a higher resolution for subsequent This paper makes first attempt employ Hausdorff derivative Laplacian operator images. In terms visual quality details, contours edges, original images noisy were sharpened by using appropriate order. Numerical results indicate that outperforms high-pass filtering, Roberts...