I-LQG control of DC-DC boost converters
Leobardo Camacho-Solorio, Aaron Sarinana-Toledo
Abstract
Leobardo Camacho-Solorio, Aaron Sarinana-Toledo
Abstract
We present a detailed design of a linear quadratic Gaussian control system with integral action (I-LQG) for a switched-mode DC-DC boost converter. Initially, identification tests were carried out to obtain an adequate linear model of the converter — using a black-box simulation model in MATLAB and Simulink. Then, we proceeded with the design of LQG and I-LQG control systems in three steps. First, we designed a linear quadratic regulator (LQR). After that, we constructed a linear quadratic estimator (LQE) of Kalman type and completed a LQG control system using the separation principle. Finally, we added integral action to the system and derived new LQR and LQE to construct the I-LQG control system. We tested the LQG and I-LQG control systems in the black-box model. These tests showed the effectiveness of the controllers for the regulation of the converter's output, for the rejection of constant perturbations, and for the attenuation of noise in measurements.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We present a detailed design of a linear quadratic Gaussian control system with integral action (I-LQG) for a switched-mode DC-DC boost converter. Initially, identification tests were carried out to obtain an adequate linear model of the converter — using a black-box simulation model in MATLAB and Simulink. Then, we proceeded with the design of LQG and I-LQG control systems in three steps. First, we designed a linear quadratic regulator (LQR). After that, we constructed a linear quadratic estimator (LQE) of Kalman type and completed a LQG control system using the separation principle. Finally, we added integral action to the system and derived new LQR and LQE to construct the I-LQG control system. We tested the LQG and I-LQG control systems in the black-box model. These tests showed the effectiveness of the controllers for the regulation of the converter's output, for the rejection of constant perturbations, and for the attenuation of noise in measurements.
Key concepts: Linear-quadratic-Gaussian control, Optimal projection equations, Linear-quadratic regulator, Control theory (sociology), Kalman filter, Controller (irrigation), MATLAB, Control system