Predictive Control of Multi-Input Switched-Capacitor DC-DC Converter with Reduced Switching Frequency

In this paper, a high step-up multi-input switched-capacitor DC-DC converter is studied for hybrid renewable energy system applications. A modified model predictive controller is proposed for this converter in order to appropriately track the reference input currents with minimum number of switching actions. In comparison to the classical PI controllers, which are widely used in this application, the proposed controller provides notable advantages such as elimination of additional modulation stages and improving the dynamic and steady state performance. Moreover, designing PI controllers and tuning their parameters is more difficult for a multi-input converter, compared to a single-input one, since the behaviors of the multi-input converter input currents are not independent. However, the conventional predictive control scheme will result in ripple components due to its variable switching frequency performance. Therefore, some modifications are made in the design procedure to overcome the aforementioned problem of the conventional predictive scheme and improve the converter efficiency. Finally, the performance of the designed controller is verified and compared with the conventional predictive control scheme through simulation in MATLB/Simulink environment.