Model updating of structure dynamics based on response surface methodology

In order to obtain an accurate structural dynamics model, a method combining the response surface and optimization was proposed. The sample points were acquired by using parametric models and optimal latin hypercube experimental design to construct the polynomial response surface model. The least squares method was used to determine the polynomial coefficients and then to test the fitting accuracy of response surface. The error between the results of the calculated response surface and the actual test was taken as the objective function. An adaptive simulated annealing algorithm was employed to optimize response surface parameters. The updated model was then obtained by substituting optimized parameters into the non-updated FEM. The GARTEUR benchmark model of the European Aviation Organization was taken as an example. The comparison of modal frequencies of non-updated and updated FEMs proves the reappearance and prediction ability of the updated FEM, and verifies the effectiveness of the model updating methodology suggested in the paper.