Innovative Soft Computing Techniques for the Evaluation of the Mechanical Stress State of Steel Plates

Thin rectangular steel plates deform locally under biaxial symmetric loads (the specimen is subjected to the load through two orthogonal test axes) with respect to their center, causing extremely complex mechanical stress distributions which are not directly measurable in order to obtain 2D representations. Strengthened by the fact that suitably induced eddy currents in these plates are locally related to mechanical stresses, in this paper, we propose an innovative approach based on eddy currents to build 2D maps capable of reconstructing the distribution of the stress state of a plate and, consequently, facilitating the evaluation of its state of health. The proposed procedure, using techniques related to AI (in particular, soft computing and fuzzy similarities), evaluates the mechanical integrity of plates in terms of an equivalent classification problem. The results obtained (with a classification percentage close to 100%) are similar to those obtained using soft computing techniques characterized by a higher computational complexity.