Virtual adaptive vector-matrix meter of the oxidizer of the combustion chamber of a gas turbine engine

DOI: 10.34759/trd-2021-121-21

Аuthors

Nikulin V. S.^*, Khizhnyakov Y. N., Storozhev S. A.

Perm National Research Polytechnic University, PNRPU, 29, Komsomolsky Prospekt, Perm, 614990, Russia

*e-mail: kalif23@yandex.ru

Abstract

Fuzzy logic methods have found wide application in control problems. However, when using the apparatus of fuzzy logic, it is necessary to sel ect models of fuzzy logical operations, being selected fr om empirical considerations. This complicates the algorithms construction for specific problems solving. Application of »classical» operations is justified only for solving simple problems and does not require a large number of rules. The purpose of the study consists in modernizing the vector-matrix approach applicable to the design of a neuro-fuzzy adaptive meter for the oxidizer of a gas turbine engine algorithm. The vector models application is being proposed, aimed at the ease of implementation, high speed and of the field application expansion.

Practical significance consists in application of the vector-matrix approach of adaptive fuzzy control in the design of the combustion chamber coefficient meter of a gas turbine engine allows computational accuracy increasing, training time reduction, the scope of application expanding at the non-deterministic objects automation in the MISO system.

Application of the matrix apparatus replaces the projections of the linguistic variable (term-set) vector with the fuzzy vectors. The main operations on fuzzy vectors are given in the works of M.A. Martsenyuk, on which basis the design of a vector fuzzy oxidizer of the combustion chamber of a gas turbine engine is considered. Relevant is application of a singleton base for fuzzifier fuzzy vectors activation, as well as the fuzzy forward and backward vector implicators application for the turboprop scalar control forming. Vector-matrix representation of the initial information is convenient when programming the state-of-the-art controllers of various objects. The disadvantage of this representation consists in the fact that it is not adaptable. Thus, the article proposes to supplement the vector part of the meter with Sugeno polynomials, which coefficients are being adjusted with the teacher by least squares method.

Keywords:

rectangular membership functions, predicates, production rules, fuzzy forward and backward implications, weighted average method, Sugeno polynomials

References

1. Rutkovskaya D., Pilin’skii M., Rutkovskii L. Neironnye seti, geneticheskie algoritmy i nechetkie sistemy (Neural networks, genetic algorithms and fuzzy systems), Moscow, Goryachaya liniya-Telekom, 2006, 452 p.
2. Takagi T., Sugeno M. Fuzzy identification of systems and applications to modeling and control, IEEE Transactions on Systems, Man, and Cybernetics, 1985, SMC-15, pp. 116-132. URL: http://dx.doi.org/10.1109/TSMC.1985.6313399
3. Gostev V.I. Proektirovanie nechetkikh regulyatorov dlya sistem avtomaticheskogo upravleniya (Design of fuzzy controllers for automatic control systems), Saint Petersburg, BKhV-Peterburg, 2011, 416 p.
4. Devyatkov V.V. Sistemy iskusstvennogo intellekta silovymi ustanovkami (Artificial intelligence systems by power plants), Moscow, Mashinostroenie, 1991, 320 p.
5. Nikulin V.S., Storozhev S.A., Abdullin D.M., Khizhnyakov Yu.N. Trudy MAI, 2020, no. 116. URL: http://trudymai.ru/eng/published.php?ID=121086. DOI: 10.34759/trd-2021-116-11
6. Storozhev S.A., Khizhnyakov Yu.N. Trudy MAI, 2021, no. 118. URL: http://trudymai.ru/eng/published.php?ID=158255. DOI 10.34759/trd-2021-118-16
7. Zadeh L.A. Outline of a New Approach to Analysis of Complex Systems and Decision Processes, 1EEE Transactions on Systems, Man and Cybernetics, 1973, no. 1, pp. 28-44.
8. Bellman R., Zade L. Prinyatie reshenii v rasplyvchatykh usloviyakh. V kn.: Voprosy analiza i protsedury prinyatiya reshenii (Decision-making in vague conditions. In the book: Issues of analysis and decision-making procedures), Moscow, Mir, 1976, pp. 172-215.
9. Zade L.A. Ponyatie lingvisticheskoi peremennoi i ego primenenie k prinyatiyu priblizhennykh reshenii (The concept of a linguistic variable and its application to making approximate decisions), Moscow, Mir, 1976, 165 p.
10. Zade L.A. Razmytye mnozhestva i ikh primenenie v raspoznavanii obrazov i klaster-analize / V kn.: Klassifikatsiya i klaster (Fuzzy sets and their application in pattern recognition and cluster analysis. In the book: Classification and cluster), Moscow, Mir, 1980, pp. 208-247.
11. Khizhnyakov Yu.N. Nechetkoe, neironnoe i gibridnoe upravlenie (Fuzzy, neural and hybrid control), Perm’, Izd-vo PNIPU, 2013, 303 p.
12. Pegat A. Nechetkoe modelirovanie i upravlenie (Fuzzy modeling and control), Moscow, BINOM. Laboratoriya znanii, 2007, 798 p.
13. Leonenkov A.V. Nechetkoe modelirovanie v srede MATLAB i FuzzyTech (Fuzzy modeling in MATLAB and FuzzyTech), Saint Petersburg, BKhV Peterburg, 2005, 736 p.
14. Yarushkina N.G. Osnovy teorii nechetkikh i gibridnykh system (Fundamentals of the theory of fuzzy and hybrid systems), Moscow, Finansy i statistika, 2004, 320 p.
15. Mamdani E.H. Application of fuzzy algorithms for the control of a simple dynamic plant, Proceedings of the Institution of Electrical Engineers, 1974, vol. 121, no.12, pp. 1585-1588.
16. Martsenyuk M.A., Seletkov I.P. Nauchno-tekhnicheskie vedomosti Sankt-Peterburgskogo gosudarstvennogo politekhnicheskogo universiteta. Informatika. Telekommunikatsii. Upravlenie, 2015, no. 2-3 (212-217), pp. 125-138.
17. Martsenyuk M.A., Polyakov V.B., Seletkov I.P. Nauchno-tekhnicheskie vedomosti Sankt-Peterburgskogo gosudarstvennogo politekhnicheskogo universiteta. Informatika. Telekommunikatsii. Upravlenie, 2012, no. 6 (162), pp. 133-141.
18. Mosolov S.V. Sidlerov D.A., Ponomarev A.A. Trudy MAI, 2012, no. 59. URL: http://trudymai.ru/eng/published.php?ID=34989
19. Baklanov A.V., Makarova G.F., Vasil’ev A.A., Nuzhdin A.A. Trudy MAI, 2018, no. 103. URL: http://trudymai.ru/eng/published.php?ID=100700
20. Chicherova E.V. Trudy MAI, 2015, no. 81. URL: http://trudymai.ru/eng/published.php?ID=57812
21. Vereshchikov D.V., Voloshin V.A., Ivashkov S.S., Vasil’ev D.V. Trudy MAI, 2018, no. 99. URL: http://trudymai.ru/eng/published.php?ID=91926

Download