A5059200545- Computational Fluid Dynamics and Aerodynamics
- Fluid Dynamics and Turbulent Flows
- Gas Dynamics and Kinetic Theory
- Model Reduction and Neural Networks
- High-Velocity Impact and Material Behavior
- Microscopic Colitis
- Inflammatory Bowel Disease
- Fatigue and fracture mechanics
- Structural Health Monitoring Techniques
Chang'an University
2024
In this work, we demonstrate that physics-informed neural networks (PINNs) tend to propagate predicted shock wave information bidirectionally in time, which does not align with the actual evolution direction of solutions hyperbolic conservation laws. This mismatch results instability and hinders reduction loss governing equations, as well initial condition by deep network. order tackle problem, simplify complexity problem constructing equivalent linear transport equations region generation....
The time series methods derived from the structural dynamic equation have been effectively applied to damage localization. current technique utilizes an autoregressive moving average with exogenous inputs (ARMAX) model, neglecting responses of certain degrees freedom, and approximating excitation noise as a (MA) process, which leads assumption constant residual variance. However, this typically reduces localization accuracy by potentially overlooking information. To address these issues,...