Data mining to online galvanic current of zinc/copper Internet atmospheric corrosion monitor

Abstract The galvanic current of a zinc/copper atmospheric corrosion monitor exposed to outdoor conditions is analysed to evaluate the corrosivity of the atmospheric environment. It is essential to develop effective and efficient models for the monitored corrosion current in order to uncover the underlying mechanism of the corrosion process. In this paper, we propose a new variable, the corrosion index, to quantify the corrosivity of the atmospheric environment. The time series of galvanic current is treated statistically to predict the corrosion index via a hidden Markov model. The prediction model performs favourably on the online corrosion data in terms of efficiency and accuracy.