A Cost Function to Assess Cracks in Simply Supported Beams with Artificial Intelligence

In this paper, we propose a model for detecting transverse cracks in simply supported beams, which can be part of more complex structural systems. The relative frequency shifts of the structure are considered a basis for damage identification. An original method developed by the authors is employed to evaluate the required modal parameters. A multi-stage optimization approach based on the rigidity loss suffered by the affected structure is employed to accurately recognize the locations of potential cracks. The outcome presented in this research shows the computational ability of the proposed model to indicate the presence and location of damages in beam-like structures.