Experimental and operational modal analysis in support of modal model updating - a test case

The introduction of new equipment on board of aircraft requires fatigue evaluations to determine maintenance intervals. This is particularly true when complex assemblies are required to be installed in aircraft areas exposed to combined stresses, most commonly shock and vibration, either random or harmonic. Beside a thorough characterization of the stress environment, an accurate equipment modal model is the basis to quantitatively determine whether the safe-life goal is met within acceptable confidence bounds, meaning the model must be updated with some source of experimental data. Traditional experimental approaches are based on Experimental Modal Analysis (EMA) techniques. With EMA, experimental modal parameters are obtained collecting structural response data resulting from the application of a deterministic, localized input stress. A further approach is based on Operational Modal Analysis (OMA) techniques, for which the input is unmeasured but assumed to be random and uniformly applied over the structure under test. Although OMA techniques require some stringent assumptions (such as input randomicity) when compared to EMA methods, they provide modal parameters for all the modes excited under real stress and constraint conditions, which might not be the case for EMA. The purpose of this paper is to compare EMA and OMA methods in determining modal parameters for a test-case for model updating, consisting in a new equipment installed in an in-service helicopter. EMA data are obtained with a Ground Vibration Test (GVT), applying a localized white noise input generated by a shaker. OMA data are collected in flight by a set of accelerometers located at the same positions of the EMA test setup. Harmonic components related to the main rotor are identified and removed by mean of a spectral statistical characterization of the input signal. Mode shapes and Frequency Response Functions derived by the application of the two techniques are quantitatively compared and the effects of changing flight conditions are determined.