电测与仪表2018，Vol.55Issue(5)：8-13,6.
基于改进深度学习的刀闸状态识别方法研究
Research of the switch state recognition method based on the improved deep learning
摘要
Abstract
The switch state recognition is essential for modern power systems,and the traditional switch state recognition methods cannot effectively solve the problem of multiple switch target interference.In order to solve this problem,a switch state recognition method based on the improved deep learning was proposed in this paper.Firstly,a spatially weighted pooling strategy was employed to improve traditional convolutional neural networks (CNNs).Secondly,the model was trained on the training database by using the improved CNNs.Thirdly,the trained model was used to detect the candidate positions of insulators and switches,and then,the exactly locations of insulators and switches were extracted via non-maximum suppression algorithm and line fitting method.Finally,the on/off state of switches was recognized by calculating length-width ratio of switch regions and connectivity between switch region and insulator regions.The experiment results show that the proposed method can accurately localize the insulators,switches and significantly improve the precision of recognizing switch state.关键词
卷积神经网络/深度学习/绝缘子检测/刀闸状态识别Key words
convolutional neural networks/deep learning/insulators location/switch state recognition分类
信息技术与安全科学引用本文复制引用
张骥,张金锋,朱能富,余娟,陈子亮..基于改进深度学习的刀闸状态识别方法研究[J].电测与仪表,2018,55(5):8-13,6.基金项目
国家电网公司科技项目(基于SOA架构的变电站一体化业务系统关键技术研究与应用) （基于SOA架构的变电站一体化业务系统关键技术研究与应用）
