Modal parameter estimation from inconsistent data sets

摘要: Most modern modal model estimation algorithms start from the observation that parameters such as resonance frequencies, damping ratios and participation factors are global for structure under test. By means of a least squares procedure, is forced on available time or frequency domain data. In many practical cases however, these data slightly to strongly inconsistent. Mass loading effects, temperature variations etc. make measured consecutively in different patches can show differing resonant frequencies. When trying fit through data, errors result by identifying multiple close poles instead one single pole near resonances. But most importantly, mode shape extraction be seriously affected residues extracted values do not correspond actual value FRF impulse response consideration. This may lead major problems postprocessing substructuring modification analysis. Also, columns matrices relating excitation tests often Shaker suspension constraints small nonlinearities indicated possible error sources. The relevance will briefly reviewed some case studies pragmatic remedies evaluated. leads methods could also useful smoothing pre-processor based impedance (substructuring, load analysis), where inconsistencies matrix inversion errors.