Taxonomic scheme synthesis to identify the states of complex systems

Functioning of complicated technological systems is connected with purpose-oriented procession of formed sensor images, which display actual system-related situations at the expense of a set of controlled factorial parameters. Introduction to these systems of high-speed digital processing means makes it possible to improve efficiency of experimental researches and expert reviews, correlation analysis of the controlled process and synthesis of adequate behavioral models, as well as development of technically feasible managerial reactions. A taxonomic scheme for assessment of systematic states is provided for development of control algorithms and hardware and software, which are characterized by a certain degree of the totality of system properties and generated managerial reactions on their basis. The state identification process presupposes the comparison of the current values of controlled parameters of the vector of factor variables with classification components of multiple dedicated taxons in the form of their median centers. Peculiarities of the probabilistic approach to the determination of median centers of taxons, as well as of the method of expert functions and fuzzy logic algorithms application have been considered. The use environment of Euclid, Hemming and Mahalanobis metrics has been analyzed as a proximity measure of the current state of the system to the aforesaid taxons.