Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
ON THE APPLICATION OF PHYSICS-INFORMED NEURAL NETWORKS IN THE MODELLING OF ROLL WAVES
Submission Author:
Bruno Fagherazzi , SP , Brazil
Co-Authors:
Bruno Fagherazzi, Valdirene Rocho, Guilherme Henrique Fiorot
Presenter: Bruno Fagherazzi
doi://10.26678/ABCM.COBEM2023.COB2023-1183
Abstract
Roll wave instability in free-surface flows is a long-studied phenomenon in fluid mechanics. Existing methods to predict roll wave properties are time-consuming, expensive, or inaccurate, prompting further exploration by the scientific community. This study evaluates the performance of Physics-Informed Neural Networks (PINNs) in modeling roll waves for laminar flows of Newtonian fluids. The objective was to determine if PINNs can successfully model roll wave behavior using a limited dataset and the governing differential equations. Seven PINNs were defined and trained using a subset of numerical results from a 2D transient flow simulation, together with the Shallow Water Equations. PINNs were used to predict wave interface heights and average streamwise velocities, which were compared to reference numerical data to assess accuracy. Results showed that PINNs accurately predicted roll wave properties such as frequency and wavelength. Wave heights and average velocities were also predicted with satisfactory accuracy, though the performance was poorer at wave peaks. Overall PINNs exhibited better performance in predicting flow height than velocity. Additionally, PINN performance decreased with higher Froude numbers, which can be attributed to underlying mathematical assumptions. This study demonstrates the potential of PINNs as mathematical tools for studying roll wave behavior, providing valuable insights and highlighting challenges in their application
Keywords
Roll Waves, Physics-informed neural networks, Free-surface flow, Scientific Machine Learning, Shallow Water Equations
