UDK 62-506.1
TO NONPARAMETRIC MODELLING OF LINEAR DYNAMIC PROCESSES OF HIGH ORDER
O. A. Ikonnikov
Siberian State Aerospace University named after academician M. F. Reshetnev 31, Krasnoyarsky Rabochy Av., Krasnoyarsk, 660014, Russian Federation E-mail: ik_ol@mail.ru
The linear dynamic systems (LDS) identification problem in conditions of nonparametric uncertainty is considered, i. e. the case, when the object priori information either insignificant or completely absent is researched. The nonpara-metric modeling algorithms based on Duhamel (convolution) integral with stochastic approximation of curve of regres-sion are constructed. This paper contains the results of nonparametric modeling of linear dynamic processes of high orders. The quality of constructed models work is registered. The method permissive to generate mathematically stable linear dynamic object of high order is offered.
identification, nonparametric modeling, dynamic object, Duhamel integral, transition function.
References
1. Ikonnikov O. A., Pervushin V. F. [The study of nonparametric models of dynamic systems]. Vestnik Sib-GAU, vol. 47, no. 1, 2013, p. 36–40. (In Russ.)
2. Medvedev A. V. [To identification of linear dynamic systems]. Algoritmy i programmy v sistemakh obrabotki eksperimental’nykh dannykh [Algorithms and programs in experimental data processing systems]. Frunze, Ilim Publ., 1975, p. 14–26.
Ikonnikov Oleg Aleksandrovich – Candidate of Engineering Science, associate professor; associate professor of the Department of System and operation analysis, Siberian State Aerospace University named after academician M. F. Reshetnev. E-mail: ik_ol@mail.ru.
