基于红外图像分割的500 kV变电站电气设备运行故障智能识别

Intelligent Identification of Electrical Equipment Operation Faults in 500 kV Substation Based on Infrared Image Segmentation

当前500 kV变电站的电气设备在运行过程中出现的故障,通常采用独立设置的智能识别节点进行识别,但识别范围较小,导致故障智能识别的响应时间较长。因此,提出了一种基于红外图像分割的500 kV变电站电气设备运行故障智能识别方法。根据识别需求和标准,采用多阶的方式对故障识别数据进行预处理,以扩大识别范围;进行识别节点的多阶部署,并在此基础上构建红外图像分割电气设备故障识别模型;采用图像增强实现故障识别处理。测试结果表明,设计的红外图像分割电气设备运行故障智能识别测试组的故障智能识别响应耗时在0.2 s以下,表明所提方法的故障识别效果更佳、识别速度更快,具有实际的应用价值。

At present,the faults of electrical equipment in 500 kV substation are usually identified by independent intelligent identification nodes,but the identification range is small,which leads to a long response time of intelligent fault identification.Therefore,an intelligent identification method of electrical equipment operation fault in 500 kV substation based on infrared image segmentation is proposed.According to the identification requirements and standards,the fault identification data are preprocessed in a multi-step way to expand the identification range.The multi-stage deployment of identification nodes is carried out,and on this basis,the fault identification model of electrical equipment for infrared image segmentation is constructed.Using image enhancement to realize fault identification.The test results show that the intelligent fault identification response time of the designed infrared image segmentation intelligent fault identification test group for electrical equipment is less than 0.2 s,which shows that the fault identification effect of the method proposed in this paper is better and the identification speed is faster,and it has practical application value.