Analysis of electrochemical noise of X70 steel in Ku'erle soil by cluster analysis

Electrochemical noise (EN) technique was employed to study the corrosion behaviors of the X70 pipeline steel in the Xinjiang Ku'erle saline soil (moisture content 1.04%) during the initial corrosion period (0-7 days). The obtained EN data were analyzed by principal component analysis (PCA), hierarchical agglomerative cluster analysis (HACA) and wavelet analysis (WA). After the PCA applied, the potential standard deviation (sigma(E)) and mean value ((E) over bar) as well as current mean value ((I) over bar) were considered as the descriptors to characterize EN distribution. Then, using sigma(E), (E) over bar, and (I) over bar as variables, three different corrosion states of the local corrosion in the soil can be classified clearly by the HACA, including instable germination state, quick development state, and stable development state. According to the WA, the different characteristics of the three corrosion stages have been further proved. The results show that the PCA-HACA method should be successful for the analysis of EN signals from the soil corrosion system.