| Autor: |
Obzherin, Yuriy E., Nikitin, Mikhail M., Sidorov, Stanislav M., Voropai, Nikolai, Stennikov, Valery, Senderov, Sergey |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2022, Vol. 2552 Issue 1, p1-8, 8p |
| Abstrakt: |
Semi-Markov and hidden Markov models are a widespread tool for modeling stochastic systems. The state codes of the semi-Markov model contain important information necessary to describe the functioning of the system, but it is not always possible to obtain it completely. In particular, it is possible to obtain some signal (information) associated with a semi-Markov process. In this case, the states of the semi-Markov model are not observed (hidden). On practice, we are dealing with a certain vector of signals obtained as a result of the system operation. Consequently, there are problems of correcting the model, analyzing the dynamics of the modeled system and predicting its states on the basis of the received vector of signals. The problems described above can be solved using the apparatus of the theory of hidden Markov models. In this work, a hidden Markov model is built on the basis of a control system semi-Markov model for hidden failures of a technological cell. It is used to solve the problems of evaluating the characteristics of a semi-Markov model, analyzing its dynamics and predicting states based on the vector of signals obtained during the system operation. The constructed hidden Markov model makes it possible to expand the possibilities of semi-Markov model using, in the presence of variables in it that cannot be observed directly. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
