Computer aided method for evaluation of failure class of materials working in creep conditions

One of the important metal science problems concerning the long-time work of the components of the pressure power installation is the evaluation of their residual life. The durability of the structural elements of the power installations (e.g. steam boilers, pipelines, and other) may be estimated with the limited certainty only. This paper presents a computer aided method for an analysis of metallographic images obtained in the scanning electron microscope (SEM). The method enables the assessment of the failure state of materials for power plants installations working in a creep condition. There are two alternative approaches presented. First one applies the self-organising artificial neural network for the classification of the metallographic images, and the second one based on the statistical analysis of the metallographic structures.