Published: Mar 19, 2024 by Daning Huang
We just published a manuscript on ArXiv for an innovative application of multi-variate Gaussian process regression to the modal analysis of spatiotemporal data, such as unsteady flow fields. Our method is particularly suitable for the cases where the data set is scarce and/or the temporal sampling is irregular; in these cases, the conventional methods such as spectral proper orthogonal decomposition do not work well. This is one of our efforts in one of our core thrusts, Scientific Data Analysis.
This work is based on the Master’s thesis of our lab member, Jiwoo Song. Good job, Jiwoo!
