Session: 34-05 Physics-based and machine learning models

Paper Number: 153777

On the Use of Machine Learning to Classify Laminar Separation Bubbles 

It is generally accepted that there exist two types of laminar separation bubbles (LSBs): short and long. The process by which a short LSB transitions to a long LSB is known as “bursting.” In this work, large-eddy simulations are used to study an LSB that develops along the suction surface of the L3FHW-LS (a new high-lift, high-work low-pressure turbine blade) at low Reynolds numbers. It is shown that the LSB bursts over a critical range of Reynolds numbers. The effect of bursting on transition, vortex shedding, and profile loss development is examined. The results of this work make clear that long LSBs are not just longer versions of short LSBs; they are phenomena unto themselves, distinct from short LSBs in terms of their vortex dynamics, time-average topology, loss footprint, etc. Typically, pressure is used to determine whether an LSB is long or short. As a rule of thumb, an LSB is said to be long if it has a “large” effect on the pressure distribution and short if it has a “small” effect on the pressure distribution— “large” and “small” being highly subjective terms. This work proposes using machine learning to make this determination more objective. A machine learning model is trained to differentiate between long and short LSBs based on pressure. The final model is demonstrated to perform well; not only does the model successfully classify the LSB over a range of Reynolds numbers, but it removes much of the ambiguity in doing so. Discriminating between long and short LSBs can be quite nuanced, especially for well-behaved airfoils at low Reynolds numbers. While the final machine learning model is not expected to be fully general, it is expected that, given sufficient training data, a more general model could be developed using the approach outlined in this work.

Presenting Author: Jared Kerestes AFRL

Presenting Author Biography: Jared Kerestes recently graduated with his PhD from Wright State University and is now working as an aerospace engineer at the AFRL in the turbine engine division.

Authors:

Jared Kerestes AFRL
Chris Marks AFRL
Mitch Wolff Wright State University
John Clark AFRL