An advanced genetic algorithm for structural damage detection

Structural damage detection is, currently, an open research topic, which can be formulated as an optimization problem. Classic techniques of optimization or heuristics, like genetic algorithms, have been used to solve it. This paper proposes to locate and quantify structural damage by using a multi-chromosome genetic algorithm (MGA). The principal characteristic of the proposed MGA is that it uses two chromosomes in order to represent a feasible solution. One chromosome is used to represent the localization of the damaged elements and the other is used to represent the damage extension. This type of representation considers that the number of damaged elements is not known at the beginning of the procedure. The objective function is formulated in terms of the changes in the natural frequencies and in the mode shapes that occur after damage. The finite element model of the structure is used to represent the undamaged and damaged conditions; however, noisy and incomplete measurements are considered. Simple and multiple damage scenarios are applied to beam and truss structures. Results show the excellent performance of the proposed methodology for detecting the damage scenarios in small structures. Simple damage scenarios are found with high confidence.