Designing expert system with artificial neural networks for in situ toughened Si₃N₄

Artificial neural networks (ANN) were applied to develop an in situ toughened Si₃N₄ expert design system in this paper. Mainly depending on experiments and tests, traditional research in materials science is costly and time-consuming. ANN belongs to a new kind of artificial intelligence technique, which can surmount difficulties in data analysis and model development, so it is suitable for non-linear and unstructured information processing in materials research. Four ANN models were developed to reciprocally predict the microstructure, sintering processing and mechanical properties of this material. Software with user-friendly graphical interface was developed with MATLAB language, which can help to easily and quickly realize the applications mentioned above.