Fine Topological Identification Method Based on the Fusion of Operation Files and Electricity Characteristic Data

It is of great significance to master the fine topology structure of the platform area-branch line, meter box and electricity meter efficiently and at low cost. Paper based on a large number of assembly branch monitoring device, measuring box equipment refinement full collection of power characteristic data, area real topology structure and distribution file data, based on the characteristic fusion data sample topology identification method, covering the province all kinds of area scene features of topology identification sample library, proposed for must level area users, the full amount of data collection of artificial intelligence data extrapolation topology identification method. According to the typical cases, the topology structure of the platform area was verified. The results show that the big data extrapolation method of multi-data feature fusion is adopted. Among the 10 scenarios verified scenes, the first level topology recognition rate is 100<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> and the second level topology recognition rate is 97.81 %. The terminal topology identification rate is 89.91 %, which has achieved remarkable results, and provides a new method for realizing the low cost and accurate identification of the topology structure of tens of millions of user stations in the province.