Author(s): R. Siddharth; Rajdip Nayek; Vamsi K. Chalamalla
Linked Author(s):
Keywords: Turbulent jet; Physics informed neural networks
Abstract: Turbulent jets are often observed in the environment as well as in engineering flows. Previous studies have focused extensively on this problem through theoretical, experimental, and computational techniques. The recent developments in machine learning (ML), combined with the exponential growth in computing resources and available data had a profound impact on a wide range of scientific and engineering domains. The aim of the present work is to explore the challenges faced while applying physics informed neural networks (PINN) to free shear flows. A novel architecture of PINN is proposed to model the Reynolds stresses in RANS equations.
Year: 2024