Neuro-symbolic model for cantilever beams damage detection

Darian M. Onchis, Gilbert-Rainer Gillich, Eduard Hogea, Cristian Tufisi

Abstract: Highlights•A method to perform non-invasive damage detection on a cantilever beam RFS dataset.•Design of a neuro-symbolic system called Logical Convolutional Neural Regressor.•Extensive validation with different methods and metrics.•Fusion of real logic and knowledge bases with deep convolutional neural networks.