In this paper the main problems for computer design of materials, which would have predefinedproperties, with the use of artificial intelligence methods are presented. The DB on inorganic compoundproperties and the system of DBs on materials for electronics with completely assessed information: phasediagram DB of material systems with semiconducting phases and DB on acousto-optical, electro-optical, andnonlinear optical properties are considered. These DBs are a source of information for data analysis. Using theDBs and artificial intelligence methods we have predicted thousands of new compounds in ternary, quaternaryand more complicated chemical systems and estimated some of their properties (crystal structure type, meltingpoint, homogeneity region etc.). The comparison of our predictions with experimental data, obtained later,showed that the average reliability of predicted inorganic compounds exceeds 80%. The perspectives ofcomputational material design with the use of artificial intelligence methods are considered.
展开▼
