Registration Number
0223U002293,
0118U000541
, R & D reports
Title
New highly effective methods for identifying control objects
popup.stage_title
Head
Silvestrov Anton M., Доктор технічних наук
Registration Date
14-02-2023
Organization
Public organization organization of veterans and graduates of the Institute of Energy Conservation and Energy Management of the National Technical University of Ukraine "Kyiv Polytechnic Institute named after Igor Sikorsky"
popup.description2
The object of research is a mathematical model, according to which the main task of managing the object is solved as best as possible.
The purpose of the work is to obtain unbiased effective estimates of parameters of mathematical models of control objects in conditions of noisy measurements of state variables.
It has been established that the identification of control objects is understood in two concepts:
1. Construction of an arbitrary mathematical model, which would reproduce with a given accuracy the display of incoming effects of the object during the weekend ("black box" problem);
2. Based on the known structure of the mathematical model, which reproduces the physics of the processes in the object, obtain unbiased and effective estimates of the physical parameters of the object (the problem of parametric identification).
The work is focused precisely on the second problem, which is successfully solved thanks to the new method of confluent analysis: obtaining unbiased effective estimates of the object parameters in noisy conditions of both output and input signals or state variables.
The method was investigated on test tasks to determine the possibility of application in real systems.
Product Description
popup.authors
Belenok Nadiya V.
Zimenkov Dmytro K.
Ladik Nataliya A.
popup.nrat_date
2023-02-14