Experiments and simulations of the structure Harmony-Gamma subjected to broadband random vibrations

T Roncen,J.-J Sinou,J.-P Lambelin

doi:10.1016/j.ymssp.2021.107849

Abstract

The structure Harmony-Gamma is a metallic assembly representative of an industrial structure for which the vibratory response is influenced by the apparition of nonlinear phenomena within two specific types of joints, the first corresponding to friction joints and the second to elastomer joints.The present study extends the previous work based on experiments and numerical simulations of the structure Harmony-Gamma subjected to harmonic vibrations [1]. More specifically, the nonlinear vibrational behaviour of the assembly subjected to random broadband excitations is studied. Broadband excitations are performed experimentally, in order to provide a first understanding of the nonlinear effect of both the friction and elastomer joints. Additionally, a global numerical methodology based on finite-element modelling and the use of the Harmonic Balance Method for the prediction of the nonlinear response of the Harmony-Gamma structure subjected to stochastic excitation is proposed.It is demonstrated that the use of a numerical model that has been validated against experimental tests can furthermore be used to achieve a refined understanding of the nonlinear phenomena and their origin.

Highlights

Knowing the vibrational response of mechanical systems has been a systematic engineering concern for decades
Experiments were conducted to investigate the nonlinear response of the system under study and, more precisely, the dependency of the structural response on random excitations
Dimensions maximum/height the experimental Frequency Response Functions (FRF) for different levels of random excitations, for two accelerometers located at the upper body and the central body

Summary

Introduction

Knowing the vibrational response of mechanical systems has been a systematic engineering concern for decades. The modelling and computing limitations of yesterday imposed simplified strategies to solve the equation of motion, and today’s computing tools are aimed at overcoming these limitations. Among the simplifying assumptions generally made in an engineering study, the two strongest limitations are linearisation of the motion equation, instead of considering the complex nonlinear dynamic behaviour of the mechanical system of interest, and the simplification of the vibrational input. There is an extremely good understanding of random signals [2] and it is possible for engineers to design mechanical systems taking into account this complex input. Very few studies nowadays propose to predict the dynamic responses of complex mechanical systems by considering the random signals.

Methods

Results

Conclusion