Efficient fatigue testing by using neural networks
KTmfk researchers Christian Witzgall and Sandro Wartzack, in collaboration with visiting professor Moh’d Sami Ashhab, present a novel strategy for fatigue testing of short fibre reinforced thermoplastics in the journal Materials. Using adaptive sampling and artificial neural networks (ANNs), the study focuses on the material PBT GF30. The research strategically pinpoints areas of high model uncertainty. Additional data collection in these areas significantly improves prediction accuracy. This innovative methodology promises to innovate fatigue life testing by allowing the creation of high quality models with reduced experimental effort. The study not only advances our understanding of short fibre reinforced thermoplastics, but also points the way to efficient advances in fatigue testing. Link to the Contribution https://doi.org/10.3390/ma17030729