Prediction of tensile elastic modulus of SiC/SiC mini-composites with the artificial neural network

This study is aimed to develop an artificial neural network (ANN)-based method for predicting the elastic modulus of Continuous SiC fiber-reinforced SiC ceramic matrix (SiC/SiC) mini-composites. The mesomechanical model with pores is established based on the microstructure of ceramic matrix mini-composites. The main factors affecting the tensile elastic modulus of mini-composites were obtained using a fidelity mesomechanical model combined with finite element analysis (FEA). Secondly, based on the main factors and the finite element simulation data, the backpropagation (BP) neural network is established to map the complex relationship between the elastic modulus and the structural parameters of the mini composite material. Finally, FEA is used to verify the rationality of the neural network. The results show that the ANN algorithm is robust in the prediction of tensile modulus of mini composite materials, and the prediction error of the neural network model using only one hidden layer and 2300 groups of data samples is less than 0.7% in the input of fiber elastic modulus, fiber volume ratio, and porosity. Therefore, the ANN model proposed in this study helps to evaluate the performance of SiC/SiC mini-composites during the design phase, shortening the design and manufacturing time.