Research on control strategy for improving stability of multi-inverter parallel system under weak grid condition.

• A generalized control strategy (composed of finite set model predictive control with delay compensation and inverse dynamic model control) is proposed, which improves the equivalent output impedance and phase margin of inverters. • The proposed control strategy can achieve order reduction of inverter's control system, which suppresses the resonance peak of the system and further improves the stability of multi-inverter parallel system. • The high robustness against various parameters of the proposed strategy have been proven through Lyapunov stability and frequency domain analysis. • The proposed control strategy can ensure that the THD value of grid connected current is less than 5% under very weak grid condition and other situations. Under weak-grid condition, the mismatch degree between grid impedance and inverter output impedance can increase, which will lead to inverter instability. Compared with conventional power grid, the parallel resonance of multi-inverter parallel system is more likely to happen. In order to improve inverter stability and suppress multiple-inverter parallel resonance under weak-grid condition, a new generalized control mode for control layer is proposed in this paper, which is based on finite set model predictive control (FCS-MPC) and inverse dynamic model control (IDM). The FCS-MPC with delay compensation is introduced into the current inner loop of the control layer, and the IDM control is introduced into the voltage outer loop, so as to make the transfer functions of current inner loop and voltage outer loop unitized and further reduce the order of the inverter's control system. Order reduction can improve the phase margin of the inverter output impedance, in other words, enhance inverter dynamic performance. Finally, the simulation model of multi-inverter parallel system is built, and by comparing with the traditional control strategies, the effectiveness of the proposed control strategy is verified. [ABSTRACT FROM AUTHOR]