Damage detection in steel frames connection based on Modal Assurance Criterion (MAC) and MAC flexibility

In this paper, an effective joint damage identification method has been proposed. Beam to columns connections modeled as a rotational mass less spring. Joint damage has been presented as a reduction in the connection rigidity or rotational stiffness factor. Because of sensitivity of mode shapes and frequencies on structural stiffness, the first mode shape and frequency of damaged frame has been used as a feature to identify damage in steel frame connections. This data is acquired by the modal analysis of damaged structure applying the finite element method (FEM). The numerical studies are carried out within the MATLAB (2016) environment, which is used for the solution of finite element problems. To identify joint damage, an optimization problem formulated in which the objective functions formulated based on Modal Assurance Criterion (MAC) and MAC flexibility. To solve the optimization problem, an effective meta-heuristic called Grey Wolf Optimizer is employed to detect damage in beam to columns connections. To evaluate the performance of presented method, three examples consists of a seven, twelve and fifteen story steel frames are chosen with two different scenarios of damage in beam to columns connections for each of them for this purpose. Results reveal that the proposed approach is effective to detect and estimate damage in steel frame connections.