Session: MSNDC-04-02 Nonlinear Dynamics of Structures

Paper Number: 94983

94983 - Escape Times in Nonlinear Oscillators - an Experimental and Numerical Comparison of Responses to White and Pink Noise 

Many engineering systems operate in a nonlinear regime, and additionally, they face uncertain loading conditions. Examples include micro-electromechanical (MEMS) devices subjected to thermal noise, nonlinear vibration energy harvesters generating energy from environmental sources, and offshore wind turbines exposed to wind and waves. Moreover, the stochastic resonance phenomenon has been observed in climatology and biology and utilized for signal processing applications. For computational convenience, the unknown random loads are commonly modeled as white noise. However, this noise model, which has an infinite energy content, is not physically representative of real situations. To account for more a realistic noise model, in this talk, the authors address the effects of pink noise on the nonlinear dynamics of mechanical systems and compare the associated responses with those associated with Gaussian white noise. Both, experimental studies and numerical simulations are presented.

A distinct characteristic of nonlinear systems is the emergence and presence of multiple coexisting stable attractors. In this setting, noise can initiate escapes from one attractor to another steady-state pattern. This change in the response characteristics can be either undesired as it can increase the stress levels in structures or beneficially enlarge the vibration amplitude in an energy harvester generating a higher power output. Thus, it is of importance to investigate the following: i) how probable an escape is and, if the escape happens and ii) the duration of the transition from one attractor to another. To this end, the response of a clamped-free beam with a tip magnet under harmonic and random excitation is studied. In the investigated parameter region the system experiences bistability. Numerical simulations as well as experiments show that pink noise reduces the escape times compared to white noise significantly. This observation implies that under pink noise escapes are more likely and, furthermore, these escapes can happen in shorter times. Hence, for applications it is important to employ the correct noise model to accurately predict impact of uncertain loadings on nonlinear dynamical systems.

Presenting Author: Thomas Breunung University of Maryland, College Park

Presenting Author Biography: Bachelor and Master of Science in Mechanical Engineering from Technical University of Darmstadt, Germany<br/><br/>Doctoral degree in Mechanical Engineering from ETH Zurich, Switzerland<br/><br/>Currently Postdoctroal researcher at the University of Maryland, College Park, US

Authors:

Thomas Breunung University of Maryland, College Park
Balakumar Balachandran University of Maryland, College Park