Augmented Linear Quadratic Tracker for Enhanced Output-Voltage Control of DC-DC Buck Converter

Abstract

This paper presents a methodic approach to synthesize a robust affine state-feedback controller to enhance the output-voltage tracking control performance of a DC-DC buck converter circuit. The proposed control scheme primarily utilizes a conventional linear-quadratic-tracker (LQT) that renders optimal control decisions based on the state-feedback of output-voltage and inductor-current. Additionally, it employs a feed-forward control term to track the time-varying reference voltage trajectories. Despite its optimality, the LQT lacks robustness in eliminating the steady-state errors and compensating the effects of bounded exogenous disturbances that are caused by high-frequency noises, load-step transients, and modeling errors. In this research, the conventional LQT is equipped with auxiliary tools to dynamically compensate the aforementioned parametric uncertainties. The existing state-space model of the system is augmented with an additional integral-of-error state-variable to eliminate the steady-state fluctuations in output-voltage response. The controller is also retrofitted with a self-tuning capacitor-current control term in order to emulate and deliver the derivative control effort. It rejects the disturbances, compensates the hysteresis effect rendered by the parasitic impedances, and improves the error convergence-rate of the response. The proposed augmented tracking controller is rigorously analyzed via experimental tests to validate its effectiveness.