PCA analysis on acoustic emission features of coated carbon steel substrate

The acoustic emission (AE) techniques were applied to monitor the electrochemical activities occurs during immersion test. The open circuit potential (OCP) measurement was recorded during acquisition of AE signal. AE acqusition were carried out and OCP values were recorded simultaneously during immersion test. Furthermore, the recorded variation number of hits was in good accordance with the variation OCP values. AE features such as energy, amplitude, hits, counts, signal strength, rise time were used to analysed the AE signal originate from the electrochemical reaction. A dimension reduction through principal component analysis (PCA) of the feature data of all groups was performed. Furthermore, PCA was used to automatically identified the correlation between AE features and three cluster were classify by k-means algorithm. The output was successfully identified the most significant features contribute in this work.