Identifying corrosion of carbon steel buried in iron ore and coal cargoes based on recurrence quantification analysis of electrochemical noise

Abstract

The effect of bulk cargo materials - iron ore and coal – on the corrosion of cargo hulls in carriers was investigated using electrochemical noise. Two reference corrosion systems were set up with the steel samples in contact with moist silica sand and immersed in NaCl solution, which generated localised corrosion and general corrosion, respectively. The electrochemical noise was measured and recurrence quantification analysis was used to extract feature variables. A random forest model using these feature variables as predictors was able to discriminate between the two reference corrosion systems. This model was successfully applied to the assessment of carbon steel corrosion in iron ore and coal. The results predicted by the model were in agreement with visual and microscopic observations of the relevant corroded steel samples. This work provides a novel analytical approach to future on-line monitoring of carrier structures in contact with bulk cargoes.