Probabilistic modeling of hybrid transition structures

[Display omitted] •The Fault Tree Analysis is adapted to investigate hybrid transition structures.•Based on distribution functions survival probabilities are calculated within FEM.•Cohesive elements are used within a stress state dominated analysis and proved.•Global survival probabilities of heterogeneous structures are calculated.•Calculations show good qualitative agreement with experiments. Within lightweight structures, often Fiber Reinforced Plastics (FRP) are used in combination with metallic materials. Most of these hybrid structures are manufactured by established methods like riveting, bolting or adhesive bonding. In order to avoid disadvantages like drilled FRP or large bond areas, the development of hybrid transition structures compatible to loads and material properties is required. To fulfill the requirements for enhanced lightweight design, novel, integral joint concepts are currently designed, dimensioned and produced by using textile, welding and casting techniques. Three concepts are under investigation which consist of different materials (titanium and Ti-alloys, glass fibers), manufacturing methods (casting, welding, textile techniques) and geometries. Various phase boundaries, materials and influences of the manufacturing processes have to be investigated that influence the structural behavior and its failure. Based on the results of Finite Element Models on the meso scale, further modeling is performed to include effects like material uncertainties and/or process influences. In this paper, a probabilistic computation procedure based on local survival probabilities and distribution functions is proposed and investigated. This approach allows to model the complex global failure behavior for each component or its interfaces as well as the whole hybrid transition zone. It also shows the interactions and consequences of certain component changes within the hybrid transition zone. First computations are carried out and compared with experiments.