We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Physics > Computational Physics

Title: Dimensionality Reduction and Reduced Order Modeling for Traveling Wave Physics

Abstract: We develop an unsupervised machine learning algorithm for the automated discovery and identification of traveling waves in spatio-temporal systems governed by partial differential equations (PDEs). Our method uses sparse regression and subspace clustering to robustly identify translational invariances that can be leveraged to build improved reduced order models (ROMs). Invariances, whether translational or rotational, are well known to compromise the ability of ROMs to produce accurate and/or low-rank representations of the spatio-temporal dynamics. However, by discovering translations in a principled way, data can be shifted into a coordinate systems where quality, low-dimensional ROMs can be constructed. This approach can be used on either numerical or experimental data with or without knowledge of the governing equations. We demonstrate our method on a variety of PDEs of increasing difficulty, taken from the field of fluid dynamics, showing the efficacy and robustness of the proposed approach.
Comments: 14 pages, 8 figures
Subjects: Computational Physics (physics.comp-ph); Dynamical Systems (math.DS)
DOI: 10.1007/s00162-020-00529-9
Cite as: arXiv:1911.00565 [physics.comp-ph]
  (or arXiv:1911.00565v2 [physics.comp-ph] for this version)

Submission history

From: Ariana Mendible [view email]
[v1] Fri, 1 Nov 2019 19:44:48 GMT (2586kb,D)
[v2] Mon, 18 May 2020 19:25:57 GMT (3180kb,D)

Link back to: arXiv, form interface, contact.