Simplified Predictive Control Strategy for Dual-Input Three-Phase Split-Source Inverter With Minimized Computational Burdens

Split-source inverters (SSIs) found vast research concerns as they utilize lower component numbers and sizes than other solutions, such as Z-source inverters (ZSIs) and quasi-ZSIs (qZSIs). Recently, multiple photovoltaic (PV) input port-based SSI has led to a further reduction of the needed components compared to single-input topologies. However, controlling multiple inputs with possible different generated powers, generating high-quality ac output voltage and current, and managing SSI’s inductor currents and capacitor voltage control represent challenging tasks for classical pulsewidth modulation (PWM) and other classical control methods. Therefore, a multiple-objective-based model predictive controller (MPC) with minimized computational burdens is proposed in this article based on two novel approaches, namely, the simplified current-based finite control set model-predictive control (SC-FCSMPC) approach and the simplified voltage-based finite control set model-predictive control (SV-FCSMPC) approach. The two proposed approaches ensure effective control of input sources during partial or complete shading in the case of two input PV sources. Moreover, the proposed approaches eliminate the need for weighting factors in the control of the cost function, simplifying the MPC design. Consequently, the two proposed MPC approaches avoid cascaded loops for controlling multiple input topologies, weighting factor adjustment procedures, and high computation burden problems. Experimental results with performance evaluations at different expected scenarios are provided in this article to confirm the superiority and applicability of the newly proposed weighting factorless MPC approaches.