Extreme value statistics of flow induced noise and vibration

Connor J. McCluskey, Stephen C. Conlon, Manton J. Guers

Research output: Contribution to conferencePaper

Abstract

Flow induced noise and vibration produces cyclic loading on structures such as wind turbines and vehicle control surfaces. This cyclic loading can often produce fatigue damage in these structures. Since the flow excitation is often random in nature, infrequent large amplitude loads are expected to occur in these applications. These large outlier loads ultimately decrease the fatigue performance of these structures. The goal of this work is to develop improved methodologies for predicting and modeling these rare events in order to establish relevant design loads. To accomplish this, the Generalized Extreme Value (GEV) model is applied to flow-induced vibration response maxima, extracted using Block Maxima methodology. The model is then compared to the Ordered Empirical Distribution (OED) to confirm model accuracy to the original data. Vibration response measurements are repeated and compared to GEV model to evaluate if extreme events were predictable.

Original languageEnglish (US)
StatePublished - Jan 1 2018
Event47th International Congress and Exposition on Noise Control Engineering: Impact of Noise Control Engineering, INTER-NOISE 2018 - Chicago, United States
Duration: Aug 26 2018Aug 29 2018

Other

Other47th International Congress and Exposition on Noise Control Engineering: Impact of Noise Control Engineering, INTER-NOISE 2018
CountryUnited States
CityChicago
Period8/26/188/29/18

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Acoustics and Ultrasonics

Cite this

McCluskey, C. J., Conlon, S. C., & Guers, M. J. (2018). Extreme value statistics of flow induced noise and vibration. Paper presented at 47th International Congress and Exposition on Noise Control Engineering: Impact of Noise Control Engineering, INTER-NOISE 2018, Chicago, United States.