Navy submarines and surface ships rely on a combination of material coatings and cathodic protection systems to protect steel tanks and enclosures from corrosion. The US Navy spends an estimated $204 million annually in direct costs [1] to locate and repair corrosion related damage on approximately 4000 tanks inspected each year. Presently, a coating health monitoring technology that relies on electrochemical sensors and stochastic models is being developed to quantify the extent and location of coating damage, while also providing users with a real-time assessment of the cathodic protection system performance. Electrochemical impedance spectroscopy (EIS) measurements from a multi-sensor network serve as the basis for assessing the coating’s health, with specific features identified and fed into a neural network model that is trained to identify and track the onset and evolution of coating defects. Previous results have demonstrated the successful performance of this system on oneand two-dimensional test articles in the laboratory. This work extends development into three-dimensional structures and to length-scales that are representative of ballast water tanks commonly found in Navy vessels.
doi: 10.12783/SHM2015/149