In this paper, a model predictive based minimum DC-link voltage control (MP-mDVC) method is proposed to enhance both the reliability and dynamic performance of grid-connected converters (GSCs). Unlike conventional voltage oriented control (VOC) methods, which typically regulate the DC-link voltage at a fixed rated value, the minimum DC-link voltage (mDVC) method introduced in the literature dynamically adjusts the DC-link voltage reference to the minimum required level based on the system's operating conditions. While the mDVC strategy can effectively enhance reliability, it may also degrade the dynamic performance of the grid-side converter. Reducing the DC-link voltage to alleviate voltage stress on components limits the available stored energy, which in turn slows the converter's response to operating changes and makes the system more susceptible to instability. To address this, the proposed method leverages the dynamic characteristics of upstream controllers in a cascaded VOC structure and introduces a predictive strategy in the outer loop to predict reference current changes and accordingly adjust the DC-link voltage setpoint. Furthermore, the inner current controller is replaced with a finite-set model predictive direct current control (MPDCC) method to eliminate the response lag associated with the current control loop. A discrete space vector modulation (DSVM) technique is also incorporated into the MPDCC to ensure constant switching frequency and maintain the quality of the grid-side current. The effectiveness of the proposed strategy is demonstrated through extensive simulations and validated via experimental testing. The results confirm that the method not only minimizes the DC-link voltage to enhance component reliability but also improves the dynamic performance of the converter.
Bizhani et al. (Thu,) studied this question.