Vibration-Based Damage Assessment in Truss Structures Using Local Frequency Change Ratio Indicator Combined with Metaheuristic Optimization Algorithms

Abstract

Structural damage assessment is of a crucial importance issue in civil, mechanical and aerospace engineering. In this study, an approach based on Local Frequency Change Ratio (LFCR) as a damage indicator and optimization techniques as a tool of damage identification is proposed. Firstly, simple and multiple damage locations are detected using the concept of LFCR. Secondly, to determine the extent of the damage accurately, an optimization problem is investigated using an objective function based on the LFCR indicator. Five recent optimization algorithms are presented, namely, Prairie Dogs Optimization (PDO), Tasmanian Devil Optimization (TDO), Artificial Ecosystem-based Optimization (AEO), Student Psychology Based Optimization (SPBO) and Flow Direction Algorithm (FDA). To test the performance of the proposed approach, two structures are studied, including a 20-Bar 2D Truss and a 28-Bar 3D Truss with different scenarios of damage. The numerical results show that the LFCR can detect and locate the damage precisely and the presented optimization techniques can define its severity accurately. Moreover, the convergence and the CPU time analysis are discussed, where a comparison between the algorithms reveals the supremacy of the SPBO over the other optimization techniques. In terms of convergence, the PDO algorithm provides less competitive outcomes.