Estimation of Corrosion of Iron Using Clustering Techniques

Corrosion is a big threat to the functionality and structural integrity of materials based on iron, demanding robust estimation strategies for timely mitigation. The conventional methods of corrosion estimation are often based on empirical models which oversimplify the complex interactions between material and environmental factors. Clustering techniques have ability of identifying the relationships and patterns within datasets offering a promising possibility for taming corrosion estimation. The unsupervised technique allows for the examination of intrinsic structures within data which reveals hidden correlations in context of corrosion. This work addresses the challenge in corrosion estimation by establishing and comparing two different clustering models using K-means and the FCM (Fuzzy C-means) techniques in MATLAB. The study initiates with an exhaustive literature survey examining existing work based on corrosion. The aim here is to estimate corrosion levels in iron on the basis of environmental factors including moisture content, oxygen level, and temperature.