Regensburg 2010 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 4: Statistical Physics (general) II

DY 4.5: Talk

Monday, March 22, 2010, 15:00–15:15, H47

Randomly driven hysteresis — •Sven Schubert and Günter Radons — Chemnitz University of Technology, D-09107 Chemnitz

Many physical and technical systems are characterized by non-trivial hysteretic behavior. Topical examples are porous materials, shape memory alloys, and magnetic nanoparticles. The fact that external fields are often entirely erratic leads to the question how hysteretic systems respond to random processes. Hence we are interested in autocorrelation or spectral properties, respectively, of Preisach hysteresis models driven by stochastic input scenarios.

Starting from a known scenario, the case of uncorrelated input signals, where long-term correlations and 1/f-noise could emerge due to hysteresis [1], we investigate the role of long-term correlations in input signals for hysteretic systems using numerical simulations. For systematic investigations, firstly, we establish a method to compute correlated input processes that allows us to determine the probability density and long-term autocorrelation decay of the input signal. Secondly, different hysteresis outputs are computed and, lastly, their correlation decay is analyzed.

Our investigation indicates that correlations in the input signal compete with hysteretically induced correlations. This results in two regimes: (1) long-time tails of input correlation survive in the output correlation or (2) we observe long-time tails due to hysteresis, as if they were induced by δ-correlated input. In the latter case long-term correlations due to hysteresis cover faster decaying input correlations.
[1] G. Radons, Phys. Rev. Lett. 100, 240602 (2008).