Structural Corrosion Health Assessment using Computational Intelligence Methods

Abstract

Corrosion greatly affects the integrity of many engineering structures, such as bridges, pipelines, nuclear reactors, and aircraft. This study provides an overview of the computational intelligence methods developed for the corrosion damage assessment of aerospace materials and structures. Specifically, cellular automata modeling of corrosion pit initiation and growth, wavelet based image processing methods for corrosion damage assessment, and artificial neural networks (ANNs) for material loss and residual strength predictions. In addition, ANN based prediction of life due to corrosion-fatigue conditions are considered and presented. Results obtained from selected computational intelligence methods are compared to the existing alternate solutions and experimental data. The results presented illustrate the feasibility of computational intelligence methods for modeling and assessing the corrosion health of aging aircraft structures and materials.