A Background/Resonant decomposition based method to predict the behavior of 2-dof aeroelastic oscillators. Issue 233 (February 2023)

• Record Type: Journal Article
• Authors: Heremans, Julien; Geuzaine, Margaux; Denoël, Vincent
• Is Part Of: Journal of wind engineering and industrial aerodynamics. Issue 233(2023)
• Publication Date: 2023-02
• Subjects: Multiple timescale spectral analysis -- Flutter -- Coupled aeroelastic oscillator -- Modal coupling -- Pitch-plunge model -- B/R decomposition; Wind-pressure -- Periodicals; Buildings -- Aerodynamics -- Periodicals; Pression du vent -- Périodiques; Constructions -- Aérodynamique -- Périodiques; Buildings -- Aerodynamics; Wind-pressure; Periodicals
• DOI: 10.1016/j.jweia.2022.105290
• Languages: English
• ISSNs: 0167-6105

Abstract

Abstract: Flutter is known as an important issue when designing flexible structures under environmental loading. The frequency-domain approach is one way to carry out this analysis under aeroelastic and buffeting forces, and consists in the determination of the variance of the response for all wind speeds of the design envelope. Each variance is classically obtained by numerical integration of the response power spectral densities (PSD), using a large number of integration points to capture the sharpness of their resonant peaks. This operation is particularly resource consuming, and considerably restricts the speed of the analysis. This paper presents an alternative semi-analytical method for integrating the PSD of a two degree-of-freedom aeroelastic model, based on the Background/Resonant decomposition of the response spectra. It focuses particularly on the integration of the covariances, that model the intermodal coupling effects in the response. The proposed method is compared to other existing Background/Resonant decomposition formulations, and recovers the Davenport's approximation when neglecting all aeroelastic effects. It is illustrated on an application under the form of a parametric study, and its accuracy is compared to classical numerical methods. The errors committed on the exact nodal variances are of order of 1%. Highlights: Extension of the Background/Resonant decomposition to coupled aeroelastic models. Asymptotic expansions are determined for the power spectral densities. The approximation speeds up the flutter analysis by accelerating the PSD integration. Exact nodal variances are reproduced with approximately 1% relative error.