In this paper, a robust voltage control is designed for a floating interleaved boost converter with high voltage gain. The proposed controller has an inner loop based on super-twisting sliding mode algorithm, who has continuous control signal and the sliding surface is defined for the inductor current. The reference current value is generated by active disturbance rejection control (ADRC) algorithm in the outer loop, based on the output voltage error. The stability of the sliding mode inner loop and the ADRC outer loop are proven using Lyapunov stability theorem and Routh-Hurwitz criteria, respectively. The robustness of the proposed controller is analyzed in depth and validated by the simulations and experimental results obtained with a 100W prototype converter.

The power semiconductor device is one of the most fragile components of the DC-DC power converters in industrial applications. In this paper, a Luenberger observer-based switch fault diagnosis method is proposed for the DC-DC interleaved boost converter (IBC) for fuel cell (FC) application. The diagnosis variable is the inductor current, which is within the main control system, without needing extra sensors. An adaptive threshold is used to avoid the effects of the converter modeling imperfection, and the wide converter operation range due to FC nonlinear volt-ampere characteristic. The proposed fast diagnosis method is robust to the converter parameter uncertainty and external disturbance. The simulation and experiment results validate the effectiveness and robustness of the proposed diagnosis method.

In this paper, a robust switch fault diagnosis method based on sliding mode observer is proposed and applied to the dc-dc multi-phase floating interleaved boost converter (FIBC). The proposed method requires only a crude converter model with nominal circuit parameters. It shows strong robustness to the circuit parameter uncertainties and disturbances. The diagnosis variables are already within the closed-loop control system, without needing extra sensors. The sampling requirement for fault diagnosis is the same level as that of the main controller. Therefore, the proposed diagnosis method can be added-on the available digital control system directly. The effectiveness and robustness of the proposed diagnosis method are validated by the simulation and experiment results. The switch open-circuit faults can be diagnosed in less than two switching periods.

To enhance the robustness and to deal with the switch fault uncertainty of floating interleaved dc-dc boost converter (FIBC), an adaptive active disturbance rejection control (ADRC) is proposed in this paper. In particular, the controller parameter b0 is adapted by obtaining approximately the system input control gain b in real time, based on the output duty cycles via digital filters. Compared with the conventional ADRC, the performance of the proposed controller is significantly improved, without the need of extra sensors. Moreover, the influences of the controller parameters on the system dynamical performance are analyzed theoretically, providing a practical guideline for controller tuning. The simulation and experiment results validate the effectiveness and the robustness of the proposed controller.

The powerful control ability of the active disturbance rejection control (ADRC) method has attracted much attention from both the academic and industrial communities. This method successfully inherits the advantages of the conventional proportional integration (PI) control method, such as error-driven way, simple structure, and low design complexity. In addition, it also absorbs the observer technique from modern control theory. All these factors make the ADRC method show a big potential to be widely adopted in practical industrial applications. To facilitate this development trend, in this work, some advancements of ADRC are discussed in detail, and three typical applications in the power electronics field are briefly introduced, such that the engineers can have a more intuitive understanding of this method and the academics could recognize some topics that still need to be further researched.