Lightning Arrester Status Evaluation based on Defect Data Mining and 1D-CNN

In order to accurately judge and analyze the operating status of distribution network lightning arresters. Unlike state analysis that predicts temporal data and sets relevant thresholds, this article uses atomic clustering and hierarchical clustering algorithms to mine lightning arrester defect data from equipment operation data to eliminate interference from related operation data and reduce the requirement of excessive data volume. Secondly, an adaptive encoding and decoding framework is used to reduce the overall data and extract features, And set it as the input of a state classification model based on one-dimensional convolutional neural networks. By effectively mining defect data of lightning arresters and using 1D-CNN models, the defect status of lightning arresters can be identified. The experimental results demonstrate the correctness and effectiveness of data mining and automatic encoder, as well as the accuracy and timeliness of the 1D-CNN state evaluation model.