VAC 05.13.01 Системный анализ, управление и обработка информации (по отраслям)
VAC 05.13.06 Автоматизация и управление технологическими процессами и производствами (по отраслям)
VAC 05.13.10 Управление в социальных и экономических системах
VAC 05.13.18 Математическое моделирование, численные методы и комплексы программ
VAC 05.13.19 Методы и системы защиты информации, информационная безопасность
UDK 519.24
GRNTI 20.01 Общие вопросы информатики
GRNTI 28.01 Общие вопросы кибернетики
GRNTI 49.01 Общие вопросы связи
GRNTI 50.01 Общие вопросы автоматики и вычислительной техники
GRNTI 82.01 Общие вопросы организации и управления
The least-squares method is widely applied when processing results received at the solution of tasks which are connected, for example, with the identification of dynamic objects or with the pattern recognition. The article considers the application of recurrent least-squares method for the parameters determination of a static object with a matrix input and a matrix output. As test input signals it is offered to use signals like a meander with single amplitude. The estimation results of the object parameters are given for a case when Gaussian noises occur at the object output. The simulation of input signals which are realizing impact on an object, and the iterative procedure of the least-squares method are executed in the Simulink environment. The blocks which are realizing the formation of the iterative procedure of parameters estimation correspond to basic formulas which are a part of the algorithm of the recurrent least-squares method. On the example of the second order object the estimates received as a result of the recurrent estimation constructed scheme are given in the graphic form. It is possible to mark that the fast convergence of the parameters estimates to basic parameter values of an object is stated. The behavior diagram of the gain coefficient which is present at the algorithm of the recurrent least-squares method is demonstrated. Testing the algorithm of the object parameters estimation was carried out using input signals like a meander with different periods. The simulation results show that the algorithm gives the good estimates of unknown parameters even in the presence of the considerable noise watched on the object output. The offered approach is supposed to be used for the parameters estimation of the higher order objects with the different parameters quantity.
identification, modeling, meander, parameters estimation, recursive least-squares method, gain coefficient
1. L'yung L. Identifikaciya sistem. Teoriya dlya pol'zovatelya. M.: Nauka, 1991. 432 s.
2. Goodwin G. C., Payne R. L. Dynamic System Identification: Experiment Design and Data Analysis. New York: Academic Press, 1977. 291 p.
3. Seydzh E. P., Uayt Ch. S. III. Optimal'noe upravlenie sistemami. M.: Radio i svyaz', 1982. 392 s.
4. Seydzh E. P., Melsa Dzh. L. Identifikaciya sistem upravleniya. M.: Nauka, 1974. 248 s.
5. Aström K. J. Maximum Likelihood and Prediction Error Methods // Automatica. 1980. V. 16. N. 5. P. 551-574.
6. Antsaklis P. J., Michel A. N. Linear systems. New York: McGraw-Hill, 1997. 685 p.
7. Chen C. T. Linear system theory and design. New York; Oxford: Oxford University Press, 1999. 334 p.
8. Fomin V. N. Rekurrentnoe ocenivanie i adaptivnaya fil'traciya. M.: Nauka, 1984. 288 s.
9. Ogarkov M. A. Metody statisticheskogo ocenivaniya parametrov sluchaynyh processov. M.: Energoatomizdat, 1990. 208 s.
10. Voskoboynikov Yu. E. Kriteriy rashodimosti i algoritm adaptacii rekurrentnogo algoritma ocenivaniya vektora sostoyaniya // Nauch. vestn. Novosib. gos. tehn. un-ta. 2015. № 3 (60). S. 7-22.
11. Cacace F., Germani A., Manes C., Setola R. A new approach to the internal positive representation of linear MIMO systems // IEEE Trans. Autom. Control. 2012. V. 57. N. 1. P. 119-134.
12. Cacace F., Germani A., Manes C. A new approach to the design internal observers for linear systems // IEEE Trans. Autom. Control. 2015. V. 60. N. 6. P. 1665-1670.
13. Fursov V. A., Goshin E. V. Adaptivnaya identifikaciya po malomu chislu nablyudeniy s kontrolem obuslovlennosti: tr. XII Vseros. sovesch. po problemam upravleniya (VSPU-2014) (Moskva, 16-19 iyunya 2014 g.). M.: In-t problem upravleniya im. V. A. Trapeznikova RAN, 2014. S. 2720-2727. URL: http://vspu2014.ipu.ru/proceedings/vspu2014.zip (data obrascheniya: 28.10.2014).
14. Azarskov V. N., Zhiteckiy L. S., Solovchuk K. Yu. Parametricheskaya identifikaciya mnogosvyaznogo staticheskogo ob'ekta v zamknutom konture upravleniya: special'nyy sluchay: tr. XII Vseros. sovesch. po problemam upravleniya (VSPU-2014) (Moskva, 16-19 iyunya 2014 g.). M.: In-t problem upravleniya im. V. A. Trapeznikova RAN, 2014. S. 2764-2776. URL: http://vspu2014.ipu.ru/proceedings/vspu2014.zip (data obrascheniya: 28.10.2014).
15. Mehra R. K. Optimal Input for Linear System Identification // IEEE Trans. Autom. Control. 1974. V. 19. N. 3. P. 192-200.
16. Voevoda A. A., Troshina G. V. Realizaciya iteracionnogo metoda naimen'shih kvadratov dlya ocenivaniya parametrov staticheskih ob'ektov v srede MATLAB // Vestn. Astrahan. gos. tehn. un-ta. Ser.: Upravlenie, vychislitel'naya tehnika i informatika. 2017. № 1. S. 28-36.
17. Troshina G. V. O metodah ocenivaniya vektora sostoyaniya v zadachah identifikacii // Sb. nauch. tr. NGTU. 2012. Vyp. 1 (67). S. 69-78.
18. Voevoda A. A., Troshina G. V. The parameters vector estimation in the steady state for the linear dynamic systems // Rroceedings of 11 International Forum on Strategic Technology (IFOST 2016) (Novosibirsk, 1-3 June 2016). Novosibirsk: Novosibirsk State Technical University, 2016. V. 1. P. 582-584.
19. Troshina G. V., Voevoda A. A. The steady-state in the parameters estimation problem for the dynamic objects // International Multi-Conference on Engineering, Computer and Information Sciences (SIBIRCON 2017) (Novosibirsk, Akademgorodok, 18-22 September 2017). Novosibirsk: 2017. P. 351-355.
20. Troshina G. V., Voevoda A. A., Patrin V. M., Simakina M. V. The object unknown parameters estimation for the 'inverted pendulum-Cart' system in the steady state // Rroceedings of the 16th Intern. Conf. of Young Specialists on Micro/Nanotechnologies and Electron Devices (EDM-2015) (Altai, Erlagol, 29 June-3 July 2015). Novosibirsk: Novosibirsk State Technical University, 2015. P. 186-188.
21. Voevoda A. A., Troshina G. V. Modelirovanie fil'tra Kalmana s obnovlennoy posledovatel'nost'yu v srede Simulink // Sb. nauch. tr. NGTU. 2015. Vyp. 2 (80). C. 7-17.
22. Troshina G. V. Modelirovanie dinamicheskih ob'ektov v srede SIMULINK. Ch. 1. // Sb. nauch. tr. NGTU. 2015. Vyp. 3 (81). S. 55-68.
23. Voevoda A. A., Troshina G. V. Rekurrentnyy metod ocenivaniya parametrov v dinamicheskom ob'ekte // Nauch. vestn. Novosib. gos. tehn. un-ta. 2016. № 4 (65). S. 7-18.
24. Troshina G. V. Ob aktivnoy identifikacii dinamicheskih ob'ektov // Sb. nauch. tr. NGTU. 2014. Vyp. 4 (78). S. 41-52.