Identification of Photovoltaic System using a Frequency-domain Approach

The proportion of renewable energy system in power system is increasing rapidly in recent years. Due to its complexity and randomness, great challenges have been bought to the stability of power system. At present, digital simulation is widely used to evaluate the operation performance of large interconnected power grid, so whether the model in digital simulation is accurate or not is of great importance. In this paper, a closed-loop frequency-domain identification method for photovoltaic system is proposed by using wide area measurement system data of power system. Firstly, a simplified photovoltaic system model for identification is established, and it is proved that the parameter identification based on measured data is a closed-loop identification problem. The frequency-domain maximum likelihood estimation method is used for the closed-loop identification. Then the consistency analysis shows that this method can keep the consistency of identification results when using sample variance. Finally, a power system of IEEE 3 generators and 9 nodes is given to verify the accuracy of the proposed method.