Improving experimental impulse-based substructuring through interface weakening

One issue of experimentally applying the impulse-based-substructuring (IBS) method is an unstable response of the assembled system due to a combination of inaccuracies in the experimentally obtained impulse response functions (IRF), the exact enforcement of interface compatibility and the cumulative nature of the conducted convolution. This paper outlines some thoughts about interface modeling for IBS and presents various methods for weakening the interface compatibility. The effectiveness of the presented methods is then tested on a numerical example with added artificial noise. The example shows that a small amount of noise can already be detrimental for the IBS method and that an effective treatment has to be found. Finally, the most promising method is tested on an experimental example, where some potential for further development of the proposed method can be seen.