Performance Comparison Between a Statistical Models a Deterministic Model, and an Artificial Neural Network Model for Predicting Damage From Pitting Corrosion

Pennsylvania State University, University Park, PA 16802 Various attempts have been made (o develop models for predicting the development of damage in metals and alloys due to pitting corrosion. These muOcIs may be divided into two cliu.ses: the cmpirieiil approach which employ.s extreme value statistics, and the deterministic approach based on perceived mechanisms for nuclcation and growth of damage. More recently. Artificial Neural Networks (ANNs), a iioniletermlnistic t)'pc of model, has been developed lo describe the progression of damage due to pitting corrosion. Wc compare the three approaches above — Statistical, deterministic, and neural network!). Our goal i.s to Illustrate the advantages and disadvantages of each approach, in order that the most reliable methods may be employed in future algorithms for predicting pitting damage functions for engineering structures. To illustrate the difficulty that We face in predicting cumulative pitting damage, we selected a set of data that was collected in the laboratory. Wc compare and contrast the three approaches by reference to this data set.