A new control method of parallel-connected inverters using receding-horizon prediction with continuous control set

Abstract

Existing control techniques of parallel-connected inverters are often complicated with multiple voltage and current loops. A large amount of tuning effort is required to ensure system stability, and the power quality is usually deteriorated due to voltage deviations. This paper proposes a new control method for parallel-connected three-phase voltage source inverters using model predictive control (MPC). High-quality voltages for local loads can be obtained. In addition, each inverter can adjust the output current to achieve proper load sharing according to the power ratings of distributed generation (DG) sources. Hot-swap capability is also achieved to facilitate the connection or disconnection operations to the common AC bus. Started with State-Space function which includes time-consuming matrix calculation can be done in PC. Thus, control gains are found and MPC can be achieved based on a digital signal processor (DSP). The proposed method is simple without complex coordinate transformation or proportional-integral (PI) regulators. The effectiveness of the proposed control strategy are verified by the test results under various scenarios, presenting promising applications in microgrids.