Zaripov O.O., Shukurova O.P., Sevinov J.U.

Abstract

Regularized algorithms for identifying linear dynamic control objects based on the concept of pseudo inversion are given. The equation of the control object is focused on solving a number of applied problems and is selected in the form of a multidimensional dynamic regression equation. To solve the equation in question, we use differentiation formulas for quadratic functionals with respect to matrix variables and regularized algorithms based on non-orthogonal factorizations and pseudo in versions of square matrices. The obtained regular algorithms contribute to increasing the accuracy of estimating the parameters of the class of dynamic control objects under consideration.

Paper Details