Forward and inverse substructuring methods for model updating of large-scale structures

Traditional model updating of large-scale structures is usually time consuming because the global structural model needs to be repeatedly re-analyzed as a whole to match the global measurements. A forward substructuring method and an inverse substructuring method are proposed in this paper for model updating of largescale structures. In the forward substructuring approach, the analytical model of the global structure is partitioned into substructures. Each substructure is analyzed independently and assembled to recover the eigensolutions and eigensensitivity of the global structure by imposing constraints at the interfaces. Within each iteration of model updating, only the modified substructures are re-analyzed and assembled with other untouched substructures to assemble the eigensolutions and eigensensitivity of the global structure. The eigensolutions of the global structure are compared with the experimental measurement for model updating purposes. In the inverse substructuring approach, the experimental properties of the global structure are disassembled into the substructural flexibility matrices by satisfying the constraints at the interfaces. Afterwards, each substructure is treated as an independent structure. The analytical model of the substructure is updated to match the substructural flexibility matrix extracted from the measurement. Both of the proposed substructuring methods are demonstrated through a laboratory-tested frame structure. The artificial damage is detected successfully, and the damage severity identified from the forward and inverse substructuring methods are consistent. Both substructuring methods are effective in model updating and damage identification.