Assessment of SCC damage by topological neural network of combined AE and EN signals

摘要

In this work a combination of Acoustic Emission (AE) and Electrochemical Noise (EN) techniques was proposed in order to investigate the SCC damage evolution of stainless steel samples. Tests were carried out using a martensitic stainless steel, UNS S 17400 (17-4PH), in an aqueous MgCl_2 environment at 100°C , with a mechanical stress applied equal to the 90% of the 0.2% yield strength. The synergic use of the two non destructive analysis technique was performed using a synchronization process. This signal conditioning is necessary in order to perform some advanced multivariate analysis techniques as the Principal Component Analysis (PCA) and data-mining techniques as the Self Organising Map (SOM). This procedure was able to identify with good reliability the electrochemical and mechanical damage phenomena occurring during the SCC phenomena.