FatAM - Fatigue in additively manufactured metals and composites: Integration of nondestructive testing and numerical modeling

The unique selling point of this multidisciplinary research project is the testing of additive manufactured (AM) metal and composite components in combination with research into the effect of detected defects on their service life. Through the use of machine learning and structural mechanics simulation, a comprehensive IO/NIO assessment of the components can thus be made. This is particularly relevant for assessing the damage tolerance of structural components with high safety requirements, e.g. in automotive and aerospace applications.
A major hurdle to date is the lack of standards for the quality assessment of AM components, especially with regard to the control of material properties, the correlation between process and structural properties (digital twin), the effects of defects and the surface quality after the removal of support structures (e.g. by herding). AM components exhibit specific process- and geometry-related defects, which have a negative impact on fatigue properties in particular. Basically, the selection of the optimal parameters for the respective material and geometry play a major role for fatigue strength.
In this project, information about the manufacturing process (in-line monitoring) and component-specific volume data (industrial computed tomography, CT) are integrated into a common database, which serves as the basis for evaluation using machine learning - e.g., to learn classification models that perform IO/NIO evaluation of the components based on selected features. The collected data is used to generate digital process twins, on the basis of which existing component defects are to be detected and their cause determined.
Damage models based on finite element analysis (FEA) will allow further assessment of existing defects on fatigue strength. FEA based on CT data is used to determine the stress distribution of structural components to determine damage safety factors and fatigue limits. With this feedback approach, the results can be used directly to optimize design criteria, such as topology-optimized components. This also applies to the combination of AM with composite materials.
Our main goal is to formulate guidelines for the optimization of AM production parameters for companies that additively manufacture metal or composite components. Reference and real components generated in the project and the evaluation of the respective damage safety factors provide an important knowledge base to enable SMEs to transfer this advanced approach directly into products - without having to perform time-consuming test series themselves. The project thus acts directly as a competence center for metal and plastic 3D printing for Austrian companies, which contributes to the location profile for all partners involved.