Analysis of hybrid CSA-DEA method for fault detection of cracked structures

Abstract

Damage is the main cause of structural failure and often occurs on structures. Maintenance in structural elements after formation of damage is not possible because it leads to the catastrophic failure of the entire structure, but the preventive steps could be taken before the failure. The prevention is done if we can get the damage location prior to the failure. This is why, in recent years, non-destructive techniques based on changes in the structural vibrations have been developed to predict the severity of the damage. In this paper, a method is proposed to locate damage by hybridising Clonal Selection Algorithm and Differential Evolution Algorithm. The inputs to the hybrid system are the relative values of the first three natural frequencies of the damaged structure and outputs are relative crack locations and crack depths. For training the hybrid system, the natural frequencies are found out using Finite Element Analysis and Experimental Analysis for different crack locations and crack depths. The test results from the proposed method are compared with finite element analysis and experimental analysis for validation.