TY - GEN A1 - Martin, Ivo A1 - Bestle, Dieter T1 - Automated eigenmode classification for airfoils in the presence of fixation uncertainties T2 - Engineering Applications of Artificial Intelligence N2 - Abstract Automated structural design optimization should take into acc ount risk of failure which depends on eigenmodes, since eigenmode shap es determine failure risk by their characteristic stress concentration pattern, as well as by their specific interaction with excitations. T hus, such a process needs to be able to identify eigenmodes with low e rror rate. This is a rather challenging task, because eigenmodes depen d on the geometry of the structure which is changing during the design process, and on boundary conditions which are not clearly defined due to uncertainties in the assembly and running conditions. The present investigation aims to find a proper classification method for eigenmod es of compressor airfoils. Specific data normalization and data depend ent initialization of a neural network using principle-component direc tions as initial weight vectors have led to the development of a class ification and decision procedure enabling automatic assignment of prop er uncertainty bands to eigenfrequencies of a specific eigenmode shape . Application to compressor airfoils of a stationary gas-turbine with hammer-foot and dove-tail roots demonstrates the high performance of t he proposed procedure. KW - Principle-component analysis Y1 - 2017 UR - https://opus4.kobv.de/opus4-UBICO/frontdoor/index/index/docId/20377 UR - https://www.sciencedirect.com/science/article/pii/S0952197617302361 SN - 0952-1976 VL - 67 SP - 187 EP - 196 ER -