Artificial Neural Network-Based Intelligent Grid Impedance Identification Method for Grid-Connected Inverter

This paper presents an intelligent grid impedance identification method for grid-connected inverter, where artificial neural network (ANN) is presented to identify time-varying grid impedance. The ANN is first trained offline by self-learning algorithm, which formulates an intelligent grid impedance identification method. Then, grid-connected inverter can identify variation of grid impedance according to output current. Simulation results are given to validate the proposed impedance identification method. The proposed impedance identification method can dynamically estimate time-varying grid impedance with good self-learning capability, so as to support the integration of renewable energies into grid.