Research on Decomposition and Identification of Wide-Band Polymorphic Signals in Active Distribution Network Based on Data Decoupling Analysis

With the large-scale renewable energy connected to the grid through power electronic equipment, the active distribution network signals show a trend of wide frequency and polymorphism, and the signal decomposition identification method based on the traditional grid operation mode is no longer applicable. For this reason, this paper proposed a decomposition and identification method of wide-band polymorphic signals in active distribution network based on data decoupling analysis. The local adaptive weighted regression filtering algorithm was used to filter the polymorphic noise in the signal, and then the improved local mean decomposition algorithm based on adaptive threshold was used for signal decomposition and parameter identification. Then, the filtering performance and decomposition identification performance of the proposed method were analyzed by simulation. Finally, the measured current data of a photovoltaic power generation unit under the active power distribution network were analyzed and compared with the analysis results of the power quality analyzer. The difference between their frequency and amplitude was not more than 0.2%, which verified the accuracy and practicality of the method applied to the signal decomposition and identification of the active power distribution network.