“Extracting Patterns from Large Datasets: An Introduction to the POD”
Thomas Bewley – Flow Control Lab, Dept of MAE, UCSD
The Proper Orthogonal Decomposition (POD) is a convenient analysis tool for the extraction of recurrent patterns from large datasets. The method has been used broadly, with some especially notable successful applications in fluid mechanics. This short talk will discuss the fundamentals of this method, as well as some of its challenges and limitations, present some recent results in the application of this method to a multi-scale flow system (a round turbulent jet), and explore some possible future applications of this method to biological systems.