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

Download:

Current browse context:

math.OC

Change to browse by:

References & Citations

Bookmark

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

Mathematics > Optimization and Control

Title: Eigenvalues of Autocovariance Matrix: A Practical Method to Identify the Koopman Eigenfrequencies

Abstract: To infer eigenvalues of the infinite-dimensional Koopman operator, we study the leading eigenvalues of the autocovariance matrix associated with a given observable of a dynamical system. For any observable $f$ for which all the time-delayed autocovariance exist, we construct a Hilbert space $\mathcal{H}_f$ and Koopman operator $\mathcal{K}$ which acts on $\mathcal{H}_f$. We prove that the leading eigenvalues of the autocovariance matrix has one-to-one correspondence with the energy of $f$ that is represented by the eigenfrequencies of $\mathcal{K}$. The proof is associated to several representation theorems of isometric operators on a Hilbert space, and the weak-mixing property of the observables represented by the continuous spectrum. We also provide an alternative proof of the weakly mixing property. The main theorem sheds light to the theoretical foundation of several semi-empirical methods, including singular spectrum analysis (SSA), data-adaptive harmonic analysis (DAHD), Hankel DMD and Hankel alternative view of Koopman analysis (HAVOK). It shows that the leading temporal empirical orthogonal functions are indeed approximations of the Koopman eigenfrequencies. A theorem-based practical methodology is then proposed to identify the Koopman eigenfrequencies from a given time series. It builds on the fact that the convergence of the renormalized eigenvalues of the Gramian matrix is a necessary and sufficient condition for the existence of Koopman eigenfrequencies. Numerical illustrating results on simple low dimensional systems and real interpolated ocean sea-surface height data are presented and discussed.
Subjects: Optimization and Control (math.OC)
Cite as: arXiv:2107.01948 [math.OC]
  (or arXiv:2107.01948v4 [math.OC] for this version)

Submission history

From: Yicun Zhen [view email]
[v1] Mon, 5 Jul 2021 11:30:16 GMT (1119kb,D)
[v2] Tue, 6 Jul 2021 10:15:55 GMT (1126kb,D)
[v3] Sun, 7 Nov 2021 18:18:25 GMT (1227kb,D)
[v4] Wed, 26 Jan 2022 14:43:09 GMT (1227kb,D)
[v5] Mon, 31 Jan 2022 12:52:44 GMT (1230kb,D)
[v6] Thu, 3 Mar 2022 10:01:07 GMT (1231kb,D)

Link back to: arXiv, form interface, contact.