NEUROFUZZY PETRI NETS FOR DEVELOPMENT AND CONTROL OF INFORMATIONANALYTICAL PROCESSES IN CYBERPHYSICAL SYSTEMS

UDC 004.822

Authors

Bobryakov Alexander V.
National Research University "MPEI", Moscow, Russia
Borisov Vadim V.
The branch of National Research University "MPEI" in Smolensk, Russia
Misnik Anton E.
BelarusianRussian university, Mogilev, Belarus
Prokopenko Sergei A.
National Research University "MPEI", Moscow, Russia

Abstract

The article is devoted to the application of the proposed variety of neuro-fuzzy Petri nets for the development and control of information- analytical processes in modern cyberphysical systems. The use of neuro-fuzzy Petri nets for modeling, monitoring and control of information-analytical processes makes it possible to accelerate the development of information and analytical processes, simplify joint work on such processes, carry out diagnostics, determine the attainability of various events within the process, their cyclicality, and also determine "bottlenecks" of such processes and discover the ways to overcome them. This approach allows to identify unnecessary complications of information- analytical processes and avoid them, significantly reduce the likelihood of creating redundant processes, reduce the number of erroneous messages about the inadmissibility of their implementation, and as a result, prevent possible errors in the development of information and analytical processes. The proposed variety of neuro-fuzzy Petri nets, including the Kwan and Cai neurons, adequately reflects the structure and dynamics of changes in the state of information and analytical processes. The nodes and transition rules of such a network are formed on the basis of a neuro-fuzzy basis of operations, the use of Kwan and Cai neurons provides an adaptive structural-parametric adjustment when changing system’s internal and external factors based on machine learning algorithms. An example of the implementation of a software-instrumental environment, including the considered approaches, is presented

Keywords

Complex cyberphysical system, Information-analytical process, Neuro-fuzzy Petri net, Kwan and Cai neuron, Ontology