Grid-connected inverter impedance acquisition method based on knowledge and data combined driving

The invention discloses a grid-connected inverter impedance acquisition method based on knowledge and data combined driving. The method comprises the following steps of S1, data acquisition: building a simulation model in a Simulink environment in Matlab, and obtaining the dq impedance big data of a grid-connected inverter under different working conditions through changing the working conditions of the grid-connected inverter, S2, model establishment: establishing a KELM model based on a variable obtained by knowledge-driven analysis as input and grid-connected inverter dq impedance as output, and optimizing a KELM input weight by using a genetic algorithm GA to obtain a GA-ELM model, and S3, error verification: establishing a semi-physical real-time simulation platform based on RT-LAB, and for a GA-ELM model, measuring and calculating the dq impedance of the grid-connected inverter under different working conditions based on a disturbance injection method. According to the method, reasonable determination o