Laser-Induced Thermal Acoustic Velocimetry with Neural Network Data Analysis

Graduate Aeronautical Laboratories
Caltech

Abstract-
Laser-Induced Thermal Acoustics (LITA) is a non-intrusive, remote, and instantaneous laser diagnostic technique for measurements of speed of sound and of the thermal diffusivity in gases simultaneously. Hence, if the gas composition is known, its temperature and density can be obtained. Two methods of LITA velocimetry are first introduced theoretically. Then, experimental results from a low-speed wind tunnel and from a supersonic nozzle are presented, which show good agreement with the theory.

The second part of the talk will focus on a new data analysis technique for LITA utilizing an artificial neural network (NN). Using a set of computed LITA signals (training set), a NN is trained to extract the speed of sound and the flow velocity from LITA signals. The NN performance is then tested with LITA signals that were not used during the training phase (validation set). This method is computationally more efficient and more robust than conventional techniques while being comparable or superior with respect to accuracy.

GALCIT Home Page

Fluids Seminar Page