Open AccessSynthesis of LQG regulator for intelligent control of the technological process of fine grinding
Author(s) -
Sabyrzhan Atanov,
А. Z. BIGALIYEVA
Publication year - 2020
Publication title -
vestnik nacionalʹnoj inženernoj akademii respubliki kazahstan
Language(s) - English
Resource type - Journals
eISSN - 2709-4707
pISSN - 2709-4693
DOI - 10.47533/2020.1606-146x.28
Subject(s) - linear quadratic gaussian control , linear quadratic regulator , control theory (sociology) , optimal projection equations , kalman filter , riccati equation , observability , controller (irrigation) , controllability , noise (video) , control engineering , engineering , optimal control , computer science , mathematics , mathematical optimization , control (management) , differential equation , mathematical analysis , artificial intelligence , agronomy , image (mathematics) , biology
The article presents the development of a linear-quadratic Gaussian controller (LQG) for intelligent control of the fine grinding technological process. The LQG regulator was designed to control the quality of mill output. The developed LQG controller takes into account external disturbances (process noise) and noise in measurements modeled as white noise with a Gaussian distribution. The controller is developed on the basis of a combination of a stationary linear quadratic controller (LQR) and estimation of the state of the Kalman filter (LQE) in the stationary state by solving the matrix Riccati equation in order to determine the feedback gain and Kalman gain. In the course of work: a mathematical model of the grinding process is built, an analysis of the frequency characteristics of the obtained model is made; the model was checked for stability, controllability, and observability; on the basis of the model, the LQG regulator was synthesized. Transient process characteristics confirm precise control. Modeling is implemented in the MATLAB environment.