Abstract: Artificial neural network (ANN) technique was applied to model the PZT based piezoelectric ceramics system. After selecting several dopants, the experimental results of 21 PZT
samples were analyzed by a BP network based on the homogenous experimental design. Calculated results indicated that the ANN model was much more accurate than multiple
nonlinear regression analysis (MNLR) model for the same set of data. Optimized formulations were also calculated and the optimized d₃₃ and K_p output values agreed well with predicted values. These results suggest that the ANN based modeling is a very
useful tool in dealing with problems with serious non-linearity encountered in the property analysis of the complicated solid solution material.

Key words: piezoelectric ceramics, artificial neural network, back-propagation algorithm, electrical properties