Structural damage detection based on swarm intelligence technique and hybrid objective function

Abstract

This paper presents a new approach for damage detection based on swarm intelligence technique and a hybrid objective function. Damage identification is treated as an undetermined inverse problem, when a limited number of measurement information is available. Then a hybrid objective function is proposed, in which the Bayesian inference and sparse technique are used. An improved swarm intelligence technique, named K-means Jaya algorithm, is developed to optimize the defined hybrid objective function to identify structural damage locations and severities. The proposed approach can be used to obtain good identification results, even when high level noise is considered and incomplete modal data is used.