An integrated micromechanical model and BP neural network for predicting elastic modulus of 3-D multi-phase and multi-layer braided composite

This research is aimed to develop an integrated methodology based on micromechanical model and neural network to predict elastic modulus of 3-D multi-phase and multi-layer (MPML) braided composite. The micromechanical model including two-scale RVC modeling and strain energy model is firstly proposed. A back propagation (BP) neural network model is then developed to map the complex non-linear relationship between microstructural parameters and elastic modulus of the composite. The 3-D braided C/C-SiC composite is used as a case study. Predictions are compared with experimentally measured response to verify the developed technique. The results show that the developed methodology performs well in predicting the properties of the complex 3-D MPML braided composite.