On the problem of obtaining consistent estimates from multi-patch modal tests

Most modem modal model estimation algorithms start from the observation that parameters such as resonance frequencies, damping ratios and modal participation factors are "global" for the structure under test and a global model is forced on the available time or frequency domain data. In reality, these data are often slightly to strongly inconsistent. First of all, modal test data are usually acquired in "patches". Mass loading effects, temperature variations etc. make that the data from different patches (or the FRF columns with multiple excitation tests) can show slightly differing resonant frequencies. This may result in important estimation errors when trying to fit a global model through these data. This may lead to major problems in postprocessing the data by modal substructuring or modification analysis. The relevance of these problems is briefly reviewed at the hand of some practical case studies and remedies are evaluated. They include selecting distributed sensor locations, analyzing data patch by patch and recombining them in a global model as well as the use of specific parameter estimation methods. Such methods could also be useful as a smoothing pre-processor for FRF based impedance methods.