Lightweight Construction - fibre reinforced plastics

 

The ambitious CO2 reduction targets in the automotive industry make the use of lightweight Fibre Reinforced Polymeric materials (FRPs) very attractive. Components made from complex FRPs must meet strict requirements on creep resistance limits, i.e. a slow progressive deformation under constant loads until failure. Determination of the necessary material models require advanced data science over large experimental databases that are often hard to find, incomplete, based on different conventions and might be related to different grades of nominally identical materials (variation of additives). Using the example of short fibre reinforced polyamide 66 (PA66GF30), BOSCH will extract [with the help of AI-powered natural language Processing (NLP) techniques] from a corpus of scientific literature several datasets (KGs) that will be structured according to Material Science ontologies.

FRAUNHOFER will provide datasets, experiments and numerical simulations, under multiaxial loads, that will be mapped to Material Sciences ontologies. Then, another BOSCH unit, as the data consumer, will be able to find and acquire these datasets via the DOME 4.0 marketplace. Moreover, the data consumer should benefit from AI facilities and interfaces of the DOME 4.0 platform, which allow to quantify the aleatoric and epistemic uncertainties in the experimental databases and thus to estimate their trustworthiness; to perform reasoning by analogy over the KG and thus to estimate the missing or implicit information.