Dynamic mode decomposition: an alternative algorithm for full-rank datasets

G. H. Nedzhibov¹

¹ Faculty of Mathematics and Informatics Shumen University Shumen 9700, Bulgaria <a href="https://orcid.org/0000-0002-7422-7369">ORCID: 0000-0002-7422-7369</a>

Applicationes Mathematicae, Tome 50 (2023) no. 1, pp. 55-65.

Résumé

Dynamic mode decomposition (DMD) is a modal decomposition technique that describes high-dimensional dynamic data using coupled spatial-temporal modes. It combines the main features of performing principal component analysis (PCA) in space, and power spectral analysis in time. The method is equation-free in the sense that it does not require knowledge of the underlying governing equations and is entirely data-driven. The purpose of this paper is to introduce a new algorithm for computing the dynamic mode decomposition in the case of full rank data. The new approach is more economical from a computational point of view, which is an advantage when working with large datasets.

Détail

DOI : 10.4064/am2465-4-2023

Keywords: dynamic mode decomposition dmd modal decomposition technique describes high dimensional dynamic using coupled spatial temporal modes combines main features performing principal component analysis pca space power spectral analysis time method equation free sense does require knowledge underlying governing equations entirely data driven purpose paper introduce algorithm computing dynamic mode decomposition full rank approach economical computational point view which advantage working large datasets