Substructural identification with incomplete measurement for damage assessment

For structural health monitoring, it is impractical to identify a large structure with full measurement due to the limited numbers of sensor and numerical difficulty in achieving convergence. The objective of the study is to develop strategies to identify structural parameters and assess structural damage based on the substructure approach with incomplete dynamic measurement. The proposed methodology is based on the Observer/Kalman Filter Identification (OKID) using input-output data via Markov parameters and Eigensystem Realization Algorithm (ERA). A solution for the identification of the structural parameters under a second-order system model from state space realization is presented. The focus is on estimating all stiffness values from the condensed stiffness matrices by model reduction using System Equivalent Reduction Expansion Process (SEREP) and Genetic Algorithm (GA) at the substructure level. The efficiency of the proposed technique is shown numerically through a multi-storey shear building subjected to random force. Results also showed that the proposed methodology is able to locate and quantify damage in a large structure with reasonable degree of accuracy.