Reshetnikov D.V. 1
Trudov A.V. 1
1 Mozhaisky Military Space Academy
The paper proposes new approaches to create assessment and forecasting the technical state of launch facilities for space purposes. Given the mathematical formulation of the general synthesis problem of the intelligent model of the system and developed methodological support its decision. The mathematical model and method of solution of the problem of determining the optimal set of informative features that can provide the validity of assessment and forecasting, and the minimum cost of their measurement is developed. A new approach to the reduction of the data measured in a numerical scale or scale of the order to the binary scale, which allows us to solve the problem of extracting knowledge from a set of heterogeneous data is proposed. This approach is based on the construction of the separating hyperplane between clusters of training data and the computation of the values of the predicate, which characterizes the belonging of the vector features of the status of one of the classes, for each informative subspaces, found by solving the problem of optimal selection of informative features The results obtained can be used as a basis for constructing closed the intelligent control system of the technical state of complex technical objects.