Application Research on Automatic Mapping of Distribution Networks Based on Big Data Technology

To solve the problem of automatic mapping of single line diagrams in distribution networks, a large amount of manual adjustment is required to meet practical requirements. This article is based on automatic mapping of a distribution network in Yunnan. The system introduces machine learning technology from the big data Spark framework to the single line graph layout algorithm, and conducts application research. Firstly, this paper introduces the automatic mapping system's Technical architecture, explaining the application environment of machine learning technology, proposing the construction of a decision tree machine learning model based on the iterative binary tree 3rd generation (ID3) algorithm. The distribution network data is used as training samples to describe the construction process of the decision tree model. Furthermore, the graph adjusted sample data is used as the training set to train the layout algorithm library. The algorithm for the layout angle of the child devices is related to the parent device. Finally, it validates the machine learning model and layout algorithm using a single line diagram data instance of the distribution network. As a result, after machine learning, the generated single line diagram runs horizontally and vertically, with balanced layering of the station building and a reasonable and beautiful layout effect, which proves big data and machine learning technology and shows the effectiveness of application in automatic layout of single line diagrams.