Modeling of microstructure and constitutive relation during superplastic deformation by fuzzy-neural network

In this paper, an adaptive fuzzy-neural network model has been established to model the microstructure evolution and constitutive relation of 15 vol.% SiCp/LY12 aluminum composite during superplastic deformation. This network integrates the learning power of neural networks with fuzzy inference systems. During the training process of the network, the back-propagation learning algorithm is applied to optimally adjust the weight coefficients of the neural network and the parameters of the fuzzy membership functions. Then, the trained network is used to predict the microstructure evolution and constitutive relation of 15 vol.% SiCp/LY12 aluminum composite during superplastic deformation. The predicted results agree very well with the experimental data of the test samples. On the basis of the good prediction ability of the proposed fuzzy-neural network, the constitutive relation and microstructure of 15 vol.% SiCp/LY12 aluminum composite under various superplastic deformation conditions have also been calculated and analyzed.