Austenitic Stainless Steel EN 1.4404 CorrosionDetection Using Classification Techniques

摘要

Different methods of classification have been used in this paper tomodel pitting corrosion behaviour of austenitic stainless steel EN 1.4404. Thismaterial was subjected to electrochemical polarization tests in aqueous environ-ment of varying chloride ion concentration (from NaC1 solutions), pH values andtemperature in order to determine values of critical pitting potentials (E_(pit)) for eachcondition tested. In this way, the classification methods employed try to simulatethe relation between E_(pit)and those various environmental parameters studied. Dif-ferent techniques have been used: Classification Trees (CT), Discriminant Analy-sis (DA), K -Nearest-Neighbours (K-NN), Backpropagation Neural Networks(BPNN) and Support Vector Machine (SVM). These models have generally beenregarded as successful. They have been able to give a good correlation betweenexperimental and predicted data. The analysis of the results becomes useful toplan improvement in the austenitic stainless steel protection and to avoid criticalconditions expositures of this material.