Research on Line and Transformer Relationship Identification Based on Limited Power Consumption Data

摘要

Line and transformer relationship identification is a procedure of determining the line to which distribution transformer. This paper suggests a Spectrum and saliency analysis identification method to eliminate challenges in line identification, with spectral analysis, is first applied to take out features with high frequency from incomplete power smart metering data and with saliency analysis extraction of noticeable features, is performed from high-frequency variations of transformers in the time domain. Similarity or Correlation analysis between line features and transformer features is used to find out transformers” line connection. The validation is done using real smart meter data collected from a distribution grid in the Midwest U.S that is made of 239 distribution transformers and the accuracy over 80% achieved. The used algorithm has been implemented using MATLAB software.