Model updating of a wing-engine structure with nonlinear connections

This paper investigates the performance of a Bayesian model updating approach using nonlinear normal modes (NNMs). The proposed approach is applied to a wing-engine structure tested in the laboratory. The structure response to a broadband input excitation is measured experimentally. The structure NNMs are then extracted from the measured data by combining a frequency-domain nonlinear subspace identification method with numerical continuation technique. An initial finite element model of the wing-engine structure is built in Matlab and its linear stiffness parameters and the coefficients of the local nonlinearities are selected as the updating parameters. The joint posterior probability distribution of the model parameters is estimated based on their prior distribution and the likelihood function which is defined as a Gaussian distribution of error function. The error function in this study is defined as the difference between model-predicted and identified NNMs at different energy levels. The calibrated model is then used to predict the structure response (NNMs and time history), taking into account model parameters uncertainties and modeling errors. The model predictions are compared with a set of experimental results that were not used in the calibration process and only considered for validation purpose.