An Identification Method of Output Impedance of Inverter Based on Nonlinear Adaptive Iteration

The output impedance information of the inverter is an important prerequisite for analyzing its grid-connected stability. In existing research, a theoretical model of the output impedance of an inverter under different conditions has been established, which is suitable for inverters whose control results and internal parameters are completely known. However, in practical engineering, especially for commercial inverters, due to the constraints of intellectual property protection and the engineering environment, it is usually difficult to obtain the control structure, algorithm, and parameters of the inverter completely and accurately, and these theoretical expressions are difficult to be directly used for sequence impedance identification. To solve this problem, this paper proposes an inverter output impedance identification method based on nonlinear adaptive iteration. The technique uses the generalized additive model to fit each measurement result, and selects each new measurement frequency point according to the fitting result error, which reduces the number of measurements in the broadband impedance identification process, improves the identification speed of the resonant peak of the impedance curve, and is suitable for the impedance identification of one or more parallel inverters.