Obtaining Corrosion Rates by Bayesian Estimation: Numerical Simulation Coupled with Data

Protecting structures from corrosion is one of the most important challenges in engineering. Cathodic protection using sacrificial anodes or impressing current from electrodes is applied to many marine structures. Prediction of the corrosion rates of structures and the design of cathodic protection systems have been traditionally based on past experience with a limited number of empirical formulae. Recently, application of numerical methods such as the boundary element method (BEM) or finite element method (FEM) to corrosion problems has been studied intensively,1-8 and these methods have become powerful tools in the study of corrosion problems.23-24 With the progress in numerical simulations, “Inverse Problems” have received a great deal of attention. The “Inverse Problem” is a research methodology pertaining to identifying unknown information from external or indirect observation utilizing a model of the system, as shown in Fig. 1. The background of inverse problems is the modeling and simulation of natural phenomena. When observations are taken of these phenomena, the observation data are used to infer knowledge about physical states. One of the most spectacular successes in the field of inverse problems was the invention of an inversion algorithm for computed tomography by Cormack (1963) and its experimental demonstration by Hounsfield (1973). The two shared the Nobel Prize in Physiology/Medicine in 1979. Applications of inverse problems arise in many fields of engineering as well. In the field of corrosion engineering, there are many issues that benefit from an inverse analysis approach. In this article, the use of numerical simulation for evaluating corrosion rates is explained with the boundary element method as an example. Then, an inverse analysis method for identifying corrosion rates or cathodic protection currents from the (easily measured) potential distribution around marine structures is introduced. This method is based on the Bayesian estimation, with the measured data and numerical simulation focused on the potential distributions around a seaside structure.