Loading Events

Application of Artificial Neural Networks to Stochastic Estimation and Jet Noise Modeling

Dr. Mark Glauser

Professor Department of Mechanical and Aerospace Engineering and Physics

Syracuse University

Thursday, February 20th, 2020

12:45 p.m.  – 2:00 p.m.

LB 272

Deep neural networks (DNNs) have shown a remarkable ability to learn complex, nonlinear relationships between sets of variables. In this talk, I will present recent work where this network architecture is applied to several different tasks relating to high-speed turbulent flows. In the first part of my talk, linear stochastic estimation (LSE) is reformulated as a machine learning problem, and the two methods are compared. Both a DNN and a LSE model are trained to estimate fluctuating pressure at a subset of locations in the near field of a Mach 0.6 jet, given the pressure measured at other locations. It is shown that DNNs exhibit a slight performance benefit over traditional LSE models on average. This approach is then applied to the supersonic multi-stream complex nozzle flow, fusing LES data from Professor Datta Gaitonde’s group at OSU, with time dependent pressure measurements in our experimentally equivalent complex multi-stream nozzle flow at SU to provide time dependent PIV measurements at Mach 1.6.  The second part of my talk will focus on the utilization of an artificial neural network (ANN) to predict the directional overall sound pressure level (OASPL) in the far field of the afore-mentioned supersonic multi-stream jet. On average, the ANN was able to predict the directional far-field OASPL within 0.75 dB, surpassing original goals. In addition to these topics, some limitations and possible extensions of the methods will be discussed.

Dr. Glauser conducts major experimental, computational and theoretical efforts to apply low-dimensional models to turbulent and transitioning flows for understanding and control for high speed aerospace applications among others. Glauser received his BS (1982) and his Ph.D. (1987) from the Department of Mechanical and Aerospace Engineering, the University at Buffalo SUNY.  Glauser has or is currently serving as: a member of the US Army Science Board (2013 – present) where he has co-chaired and been a member of a range of studies; a member of the NASA Langley Fundamental Aerodynamics Peer Review Panel (2014, 2009); a member of the ARO Mechanics program oversight board (2017 – present); Associate Editor, AIAA Journal (2007 – 2016); Program Manager for the Turbulence and Internal Flows Program at the US Air Force Office of Scientific Research (AFOSR) from 1996-1999. Glauser has obtained more than 12.7 Million dollars in research funding as PI or Co-PI. His current funding as PI is approximately 1 Million dollars from AFOSR and Spectral Energies LLC/AFRL/Lockheed Glauser has published more than 120 peer-reviewed publications and conference proceedings and has presented more than 100 invited presentations and keynote talks worldwide.