03.1 - Aerodynamics – CFD Methods and Validation
APPLICATION OF GENETIC PROGRAMMING AND ARTIFICIAL NEURAL NETWORK APPROACHES FOR RECONSTRUCTION OF TURBULENT JET FLOW FIELDS
V. Gryazev, United Kingdom; U. Armani, United Kingdom; A.R. Murali, France; A.P. Markesteijn, United Kingdom; S.A. Karabasov, United Kingdom; V. Toropov, United Kingdom; S.E. Naghibi, United Kingdom; V. Riabov, Japan
The high-speed jets considered in this research correspond to the NASA Small Hot Jet Acoustic Rig (SHJAR) experiment at acoustic Mach numbers 0.5 and 0.9. The Spectral Proper Orthogonal Decomposition (SPOD) method is applied to compress the original LES data. The use of SPOD enabled separating spatial and temporal coherent structures which spread over a wide range of frequencies and can be considered as an extension of Proper Orthogonal Decomposition (POD) method to time-resolved data. The resulting time-dependent coefficients of the SPOD modes are reconstructed using machine learning methods.