A5003061383- Model Reduction and Neural Networks
- Image Processing Techniques and Applications
- Meteorological Phenomena and Simulations
- Advanced Optical Sensing Technologies
- Advanced Fluorescence Microscopy Techniques
- Fluid Dynamics and Turbulent Flows
- Heat Transfer and Optimization
- Image and Signal Denoising Methods
- Electric Motor Design and Analysis
- Magnetic and Electromagnetic Effects
- Magnetic Bearings and Levitation Dynamics
- Heat Transfer and Boiling Studies
Arizona State University
2023
University of Hawaiʻi at Mānoa
2022
This study employs physics-informed neural networks (PINNs) to reconstruct multiple flow fields in a transient natural convection system solely based on instantaneous temperature data at an arbitrary moment. Transient problems present reconstruction challenges due the temporal variability of across different phases. In general, large errors are observed during incipient phase, while quasi-steady phase exhibits relatively smaller errors, reduced by factor 2–4. We hypothesize that vary phases...
Magnetic levitation in general is relevant to applications which require fine control of motion, vibration, force and torque. One common method for magnetic use four iron core coils connected pairs, two Hall effect sensors, an outer ring permanent magnets levitate disk from below. This uses only feedback loops stabilize rotation translation the magnet together both horizontal directions, where vertical passively stable its axis yaw left uncontrolled rotate freely. The advantageous features...
This study employs physics-informed neural networks (PINNs) to reconstruct multiple flow fields in a transient natural convection system solely based on instantaneous temperature data at an arbitrary moment. Transient problems present reconstruction challenges due the temporal variability of across different phases. In general, large errors are observed during incipient phase, while quasi-steady phase exhibits relatively smaller errors, reduced by factor 2 4. We hypothesize that vary phases...
Reconstructing fields from sparsely observed data is an ill-posed problem that arises in many engineering and science applications. Here, we investigate the use of physics-informed neural networks (PINNs) to reconstruct complete temperature, velocity pressure sparse noisy experimental temperature obtained through single-color laser-induced fluorescence (LIF). The PINNs are applied laminar mixed convection system, a complex but fundamentally important phenomenon characterized by simultaneous...