The nature of statistical learning theory
The nature of statistical learning theory
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
| Hi-index | 0.00 |
In this paper we utilize Support Vector Machines to predict the degradation of the mechanical properties, due to surface corrosion, of the Al 2024-T3 aluminum alloy used in the aircraft industry Pre-corroded surfaces from Al 2024-T3 tensile specimens for various exposure times to EXCO solution were scanned and analyzed using image processing techniques The generated pitting morphology and individual characteristics were measured and quantified for the different exposure times of the alloy The pre-corroded specimens were then tensile tested and the residual mechanical properties were evaluated Several pitting characteristics were directly correlated to the degree of degradation of the tensile mechanical properties The support vector machine models were trained by taking as inputs all the pitting characteristics of each corroded surface to predict the residual mechanical properties of the 2024-T3 alloy The results indicate that the proposed approach constitutes a robust methodology for accurately predicting the degradation of the mechanical properties of the material.