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

Download:

Current browse context:

cs.CE

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

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

Computer Science > Computational Engineering, Finance, and Science

Title: Data-driven nonintrusive reduced order modeling for dynamical systems with moving boundaries using Gaussian process regression

Abstract: We present a data-driven nonintrusive model order reduction method for dynamical systems with moving boundaries. The proposed method draws on the proper orthogonal decomposition, Gaussian process regression, and moving least squares interpolation. It combines several attributes that are not simultaneously satisfied in the existing model order reduction methods for dynamical systems with moving boundaries. Specifically, the method requires only snapshot data of state variables at discrete time instances and the parameters that characterize the boundaries, but not further knowledge of the full-order model and the underlying governing equations. The dynamical systems can be generally nonlinear. The movements of boundaries are not limited to prescribed or periodic motions but can be free motions. In addition, we numerically investigate the ability of the reduced order model constructed by the proposed method to forecast the full-order solutions for future times beyond the range of snapshot data. The error analysis for the proposed reduced order modeling and the criteria to determine the furthest forecast time are also provided. Through numerical experiments, we assess the accuracy and efficiency of the proposed method in several benchmark problems. The snapshot data used to construct and validate the reduced order model are from analytical/numerical solutions and experimental measurements.
Subjects: Computational Engineering, Finance, and Science (cs.CE); Dynamical Systems (math.DS)
Journal reference: Computer Methods in Applied Mechanics and Engineering, (2021) 373, 113495
DOI: 10.1016/j.cma.2020.113495
Cite as: arXiv:2103.09790 [cs.CE]
  (or arXiv:2103.09790v1 [cs.CE] for this version)

Submission history

From: Zhan Ma [view email]
[v1] Wed, 17 Mar 2021 17:23:50 GMT (3823kb,D)

Link back to: arXiv, form interface, contact.