MODIFICATION OF THE METHOD OF LEAST SQUARES

A. M. Silvestrov, О. М. M'yakshylo, G. I. Kryvoboka

Abstract


The article deals with the problems of using a focused multilevel information system in optimal management. In order to make such a system more efficient, it is proposed to modify the least-squares method, which provides unbiased estimates of the parameters in the real situation of noisy measurements of input and output signals of the primary converters. Parameters are estimated using the proposed method, which ensures smoothing of external influences of the model under study on the results. The efficiency of the considered method is confirmed by comparison with the method of least squares.


Keywords


Parametric identification; ordinary least squares; unbiasedness; estimation efficiency; modification of ordinary least squares; weight function

References


Methods of classical and modern theory of automatic control: Textbook in 3 volumes, under. ed. N. D. Egupova. Moscow: From the Moscow State Technical University. NOT. Bauman, 2000, 1000 p. (in Russian)

M. Ya. Ostroverkhov and A. M. Silvestrov, Methods of study of electrical systems and complexes, Kyiv: TALKOM, 2019, 300 p.(in Ukrainian)

A. N. Silvestrov and P. I. Chynaev, Identification and optimization of automatic systems, Moscow, Energoatomizdat, 1987, 200 p. (in Russian).

P. Eikhoff, Fundamentals of identification of control systems, Moscow: Mir, 1975, 683 p. (in Russian)

A. N. Tikhonov and V. Ya. Arsenin, Methods for solving ill-posed problems, Moscow: Nauka, 1979, 286 p. (in Russian).

A. G. Ivakhnenko, Long-term forecasting and management of complex systems, Kyiv: Technica, 1975, 312 p. (in Russian).

A. N. Kolmogorov and S. V. Fomin, Elements of the theories of functions and functional analysis. Moscow: Nauka, 1972 (in Russian).


Full Text: PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.