Traditional proportional-integral-based (PI-based) voltage balancing control (VBC) for single-phase cascaded Hbridge rectifiers (CHBR) suffer from unsatisfactory dynamic performance. An H_∞-VBC approach is thus proposed in this paper to address this issue. The relationship between difference of dc-side voltage square and power is deduced, which is used to construct a new feedback closed-loop for VBC system. Compared with PI-based control method, the merit of H_∞ approach is tracking performance and disturbance suppression can be quantitatively solved through sensitivity weighting function. Considering external disturbance on control output, an H_∞-VBC strategy is designed for specified frequency bands. With the proposed method, dynamic performance of voltage balancing is improved. An experimental prototype is developed, and experiments are performed to verify effectiveness of the H_∞-VBC strategy.

For $dq$ control strategies in single-phase pulse width modulation (PWM) converters, the $\beta$-axis current must be created by imaginary axis current estimation (IACE) methods. The estimated error of the $\beta$-axis current during the transient process causes $d$- $q$ axis current loops to be incompletely decoupled, thereby affecting the dynamic performance of the current loop. The second-order generalized integrator (SOGI) method suffers from slow dynamic response. The fictive-axis emulation (FAE) method provides fast dynamic response but it is sensitive to circuit parameters. A reference-input (RI)-based IACE method is proposed to overcome the above shortcomings. According to the characteristic that the $\beta$-axis current loop has no transient process, the $\beta$-axis current is estimated by the $d$- $q$ axis reference inputs. This is equivalent to introducing the $\beta$-axis reference input as a feedforward term into the $d$- $q$ axis current loop, so the parameter sensitivity problem is solved, and the parameter tuning is not needed. The proposed method can maintain good steady-state performance and significantly improve the dynamic performance of the current loop. Furthermore, it is straightforward and can be easily implemented in digital controllers. Comprehensive hardware-in-the-loop (HIL) experimental comparisons with the SOGI and FAE methods have been conducted to verify the correctness and effectiveness of the proposed RI-based IACE method.