A Neural Approach to Crack Identification in Shafts using Wave Propagation Signals

The presence of crack-like defects in mechanical and structural elements produces failures during their service life that in some cases can be catastrophic. The increasingly importance of problems related with safety and costs derived from these catastrophic failures, have pushed the researchers in the field of damage detection, to look for and develop methods for the detection and identification of defects. In this work, the wave propagation in a cracked shaft has been numerically analyzed and numerical results have been use to identify the crack through a very well known technology as artificial neural network (ANN) as the first step to establish a non-destructive method for the detection and identification of cracks using a dynamic test. Cracks with increasingly depths and shape fronts have been considered and studied. The results obtained in this work would allow the development of an on-line method for damage detection and identification for cracked mechanical elements using an easy and portable dynamic testing device.