Data-Driven Adaptive Extended State Observer-Based Model-Free Disturbance Rejection Control for DC–DC Converters

This article proposes a novel data-driven model-free disturbance rejection control (DMFC) architecture for the output voltage regulation of dc–dc converters without model priori information. Each dc–dc converter is subject to circuit uncertainties and external disturbances in the controller design. To address these challenges, a DMFC scheme based on a data-driven adaptive extended state observer (DAESO) is proposed to implement desired output voltage regulation for the dc–dc converters in the presence of complex disturbances. Specifically, a DAESO is designed by incorporating a data memory stack and a parameter learning law to estimate the unknown control input gain and lumped disturbance. A DMFC law based on switching gain and estimated information is then employed to generate the desired control signal for the dc–dc converters. The Lyapunov stability of the error dynamic system is proven. A significant advantage of the proposed scheme is that the circuit uncertainties, external disturbances, and unknown control input gain of the dc–dc converters can be simultaneously recovered without requiring any model priori information. In practical application, we apply the proposed DMFC scheme to the receiver-side dc–dc converter within a wireless charging system (WCS) to improve the output voltage regulation performance. Experimental results substantiate the effectiveness of the designed DMFC architecture for the dc–dc converters in the WCS, achieving constant output voltage regulation with load independent.