Problematic aspects of building a knowledge base in the tasks of forecasting of the durability of corrosive structures


  • Larysa Korotka


fuzzy knowledge base, rule base, fuzzy inference, forecasting of durability, corroding structures


The paper discusses the main stages of building a fuzzy knowledge base in predicting forecasting of the durability of corrosive constructions. When solving this class of problems, it is necessary to formalize incomplete information (in particular, on the parameters of an aggressive environment) and structure multidimensional data arrays that are obtained as a result of the repeated solution of these problems. In this capacity, it is proposed to use fuzzy models of knowledge representation, which, in turn, are formalized in the form of fuzzy knowledge bases. The problematic aspects of their construction are considered, including: determination of the parameters of membership functions; methods for their building; number of terms-sets of a linguistic variable; assignment of semantic and syntactic rules; number of rules of fuzzy knowledge base; establishing the completeness of a fuzzy model and the linguistic completeness of the base of rules for model; accuracy of fuzzy model; coherence and consistency of rule base.


1. Garmider L.D., Taranenko I.V., Korotka L.I., Begma P.O. Methodological approach to labor potential assessment based on the use of fuzzy sets theory. Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu, № 6 – 2019. 144-149 p. (
2. Korotka L.I., Korotka Y.A. The use of elements of computational intelligence in problems of forecasting of corroding constructions durability. Mathematical and computer modelling. Series: Technical sciences – 2017. Issue 16. – P 64-71.
3. Pegat A. Fuzzy modeling and control. Springer Studies in Fuzziness and Soft Computing, 2001. 728 p.
4. Zelentsov D.G., Korotkaya, L.I. Tehnologii vyichislitelnogo intellekta v zadachah modelirovaniya dinamicheskih sistem: monografiya [Technologies of Computational Intelligence in Tasks of Dynamic Systems Modeling: Monograph], Balans-Klub, Dnepr, 2018. 178 p. (