LIBS chemical image analysis for fatigue estimation in industrial materials

The extreme conditions industrial materials are usually exposed induce metallurgical modifications which can lead to a premature fatigue causing cracks. In this paper, we present a set of methodologies for the automation of an in-situ fatigue estimation system. The modifications considered are detectable at the material´s surface through a spatial chemical analysis process that in this system is performed by laser spectroscopy (LIBS). For the chemical analysis automation a set of signal processing techniques for the LIBS spectrum analysis is presented. The result of the chemical analysis in different points is displayed on a set of images. The estimation process is based on the application of digital image processing techniques and artificial intelligence techniques on the obtained image. This system was developed in the framework of the EC project - LIBSGrain which designed, built and tested the equipment on a chemical plant. The results obtained with the system are also presented in this paper.