Zara, A.Belaidi, I.Oulad Brahim, A.Khatir, S.Capozucca, R.Abdel Wahab, M.2023-04-032023-04-032023978-981194834-321954356DOI 10.1007/978-981-19-4835-0_7https://link.springer.com/chapter/10.1007/978-981-19-4835-0_7https://dspace.univ-boumerdes.dz/handle/123456789/11270Recently, the development of optimization techniques based on artificial neural network (ANN) has shown considerable progress in the field of damage identification in composite structures, due to their simplicity, greater precision, and lower computational time compared to non-destructive testing methods (NDT). In our work, a finite element model is developed using ABAQUS software to validate the vibratory behaviors of experimental tests. Then, based on digital data extracted from a calibrated model of the damaged CFRP cantilever specimens, we used a novel artificial neural network approach to detect and identify notch depth in carbon fiber reinforced polymer (CFRP) beam based on modal analysis. The results show that ANN based on natural frequencies can be used to identify notch depth with good accuracy in composite structuresenArtificial neural networks (ANN)Carbon fiber reinforced polymer (CFRP)FEMNotch depth identificationOther