A modelling approach for the vibroacoustic behaviour of aluminium extrusions used in railway vehicles

Modern railway vehicles are often constructed from double walled aluminium extrusions, which give a stiff, light construction. However, the acoustic performance of such panels is less satisfactory, with the airborne sound transmission being considerably worse than the mass law for the equivalent simple panel. To compensate for this, vehicle manufacturers are forced to add treatments such as damping layers, absorptive layers and floating floors. Moreover, a model for extruded panels that is both simple and reliable is required to assist in the early stages of design. An statistical energy analysis (SEA) model to predict the vibroacoustic behaviour of aluminium extrusions is presented here. An extruded panel is represented by a single global mode subsystem and three subsystems representing local modes of the various strips which occur for frequencies typically above 500 Hz. An approximate model for the modal density of extruded panels is developed and this is verified using an FE model. The coupling between global and local modes is approximated with the coupling between a travelling global wave and uncorrelated local waves. This model enables the response difference across the panels to be predicted. For the coupling with air, the average radiation efficiency of a baffled extruded panel is modelled in terms of the contributions from global and local modes. Experimental studies of a sample extruded panel have also been carried out. The vibration of an extruded panel under mechanical excitation is measured for various force positions and the vibration distribution over the panel is obtained in detail. The radiation efficiencies of a free extruded panel have also been measured. The complete SEA model of a panel is finally used to predict the response of the extruded panel under mechanical and acoustic excitations. Especially for mechanical excitation, the proposed SEA model gives a good prediction compared with the measurement results.