A micromechanics and machine learning coupled approach for failure prediction of unidirectional CFRP composites under triaxial loading: A preliminary study

This study presents a hybrid method based on artificial neural network (ANN) and micro-mechanics for the failure prediction of IM7/8552 unidirectional (UD) composite lamina under triaxial loading. The ANN is trained offline by numerical data from high-fidelity micromechanics-based representative volume element (RVE) model using finite (FEM). RVE adopts identified constituent parameters inverse analysis calibrated interface strengths form uniaxial biaxial tests. A loading strategy proposed to obtain points sliced surfaces whilst maintaining constant stress at different surfaces. It has been found that algorithm robust in UD when subjected conditions, with over 97.5% accuracy being achieved shallow , where only two hidden layers 560 samples are used. predicted 3D surface an elliptical paraboloid shape shows extremely high strength compression. approach could be used inform modification existing criteria propose ANN-based criteria.