Optimization method for uncertain boundary constraint of structure dynamical model based on test modal parameter

Accurate finite element is the premise for modal analysis of structure. For the Structure containing complex connecting mechanism, highly accurate natural frequencies and mode shapes can usually be obtained by ground modal test, but the error of modal mass is relatively large. The finite element model can accurately describe the geometry of the structural and material information, but it is difficult to give accurate boundary constraints. The combination of modal testing and finite element modeling is used in this paper. The paper develop an optimization method for uncertain boundary constraint of structure dynamical model based on test modeling; the method uses multi-objective genetic algorithm (GA), the initial finite element model is optimized by replacing structural elastic boundary constraint with the norm of natural frequencies and mode shapes. The boundary constraint optimization of a peer-section beam is used to verify the feasibility of the proposed method. And then a control surface model with modal test data is used to verify the applicability of the presented method. The results indicate preliminarily that the natural frequencies and mode shapes obtained with the optimization method agreed respectively with those obtained with modal test very well. In addition, accurate modal mass obtained with optimization can correctly revise the modal mass obtained with modal test.