Generalized play hysteresis operators in limits of fast-slow systems

Christian Kuehn, Christian Münch

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

Hysteresis operators appear in many applications, such as elasto-plasticity and micromagnetics, and can be used for a wider class of systems, where rate-independent memory plays a role. A natural approximation for systems of evolution equations with hysteresis operators are fast-slow dynamical systems, which-in their used approximation form-do not involve any memory effects. Hence, viewing differential equations with hysteresis operators in the nonlinearity as a limit of approximating fast-slow dynamics involves subtle limit procedures. In this paper, we give a proof of Netushil's "observation" that broad classes of planar fast-slow systems with a two-dimensional critical manifold are expected to yield generalized play operators in the singular limit. We provide two proofs of this "observation" based upon the fast-slow systems paradigm of decomposition into subsystems. One proof strategy employs suitable convergence in function spaces, while the other considers a geometric strategy via local linearization and patching adapted originally from problems in stochastic analysis. We also provide an illustration of our results in the context of oscillations in forced planar nonautonomous fast-slow systems. The study of this example also strongly suggests that new canard-type mechanisms can occur for two-dimensional critical manifolds in planar systems.

Original languageEnglish
Pages (from-to)1650-1685
Number of pages36
JournalSIAM Journal on Applied Dynamical Systems
Volume16
Issue number3
DOIs
StatePublished - 2017

Keywords

  • Canard
  • Fast-slow system
  • Generalized play
  • Hysteresis operator
  • Multiple time scale dynamics
  • Netushil's observation

Fingerprint

Dive into the research topics of 'Generalized play hysteresis operators in limits of fast-slow systems'. Together they form a unique fingerprint.

Cite this