A5072876977- Fluid Dynamics Simulations and Interactions
- Numerical methods in engineering
- Dam Engineering and Safety
- Hydraulic Fracturing and Reservoir Analysis
- Bone health and osteoporosis research
- Vibration and Dynamic Analysis
- Seismic Imaging and Inversion Techniques
- Hydraulic flow and structures
- Model Reduction and Neural Networks
- Diabetes Management and Education
Hohai University
2023-2024
Eli Lilly (United States)
2012
Lilly (China)
2009
Abstract A Physics-Informed Neural Network (PINN) provides a distinct advantage by synergizing neural networks' capabilities with the problem's governing physical laws. In this study, we introduce an innovative approach for solving seepage problems utilizing PINN, harnessing of Deep Networks (DNNs) to approximate hydraulic head distributions in analysis. To effectively train PINN model, comprehensive loss function comprising three components: one evaluating differential operators, another...
A Physics-Informed Neural Network (PINN) provides a distinct advantage by synergizing neural networks' capabilities with the problem's governing physical laws. In this study, we introduce an innovative approach for solving seepage problems utilizing PINN, harnessing of Deep Networks (DNNs) to approximate hydraulic head distributions in analysis. To effectively train PINN model, comprehensive loss function comprising three components: one evaluating differential operators, another assessing...
This paper presents a PSBFEM approach that integrates the quadtree mesh generation technique based on digital images for solving seepage problems. The quantitative representation of distribution geometrical information and material parameters is achieved by utilizing color intensity each pixel, which can then be applied to analysis. presented method addresses issue hanging nodes treating them as polygonal element. We validate proposed three benchmark Results show image-based domain...