Naslov
Spectral properties of the stochastic Koopman operator and its numerical approximations

Autori
Črnjarić-Žic, Nelida ; Maćešić, Senka ; Mezić, Igor

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
ApplMath2018 - book of abstracts / - , 2018, 23-23

Skup
Ninth Conference on Applied Mathematics and Scientific Computing (ApplMath18)

Mjesto i datum
Šibenik, Hrvatska, 17.09.2018. - 20.09.2018

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
stochastic Koopman operator, DMD algorithm, Hankel matrix

Sažetak
A generalization of the Koopman operator framework, originally developed for de- terministic dynamical systems, to discrete and continuous time random dynamical systems results with the stochastic Koopman operators. The study of the spectral objects of these operators (Koopman eigenvalues, eigenfunctions and modes) could be useful in analysing the behaviour of the considered random dynamical system. We provide the results that characterize the spectrum and the eigenfunctions of the stochastic Koopman operators for the particular classes of random and stochastic dynamical systems. The numerical approximations of the eigenvalues and eigen- functions of the stochastic Koopman operator are computed by using DMD RRR algorithm. Its behaviour in the stochastic framework is explored on several test examples. Furthermore, we introduce the isostables and isochrones associated to the random dynamical systems and compute their numerical approximations on the chosen example.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Temeljne tehničke znanosti

Napomena
Rad je sufinanciran DARPA projektom i potporom Sveučilišta u Rijeci.

POVEZANOST RADA