Sparse structure preserving finite element model updating problem for gyroscopic systems

Abstract

An important yet difficult aspect of the finite element model updating problem is to preserve the original inherent structures in the updated model. The analytical matrices resulting from the discretization of distributed parameter systems using finite element techniques are often symmetric and sparse. In this paper, an iterative method is proposed for solving the constrained optimization problem (COP) of gyroscopic systems, that is, finding the optimal updated matrices M̂,Ĝ and K̂ that satisfy the constraints of the characteristic equation, symmetry (or skew-symmetry) and sparsity. The proposed method is proved to be convergent and numerical examples verify the effectiveness of the method.