基于三相自搜寻比较的电气设备过热故障识别  

Overheating fault identification of electrical equipment based on three-phase self-searching and comparison

作　　者：梅开锋 朱超 MEI Kaifeng;ZHU Chao(Wuhe County Power Supply Company of State Grid Anhui Electric Power Co.,Ltd.,Bengbu 233300,China)

机构地区：[1]国网安徽省电力有限公司五河县供电公司,安徽蚌埠233300

出　　处：《电子设计工程》2023年第23期22-25,30,共5页Electronic Design Engineering

摘　　要：针对现有方法存在电气设备过热故障识别准确率较低的问题,提出基于三相自搜寻比较的电气设备过热故障识别方法。预处理电气设备红外图像,通过归一化处理获取图像的局部极大值和极小值,并将电气设备红外图像转换为灰度图像;采用激光传感器获取电气设备图像的边缘特征,通过三相自搜寻比较方法将电气设备红外图像划分为大小和姿态相似的部分,判断对应图像块是否存在异常升温,判定电气设备是否出现热故障,实现电气设备过热故障识别。实验结果表明,所提方法的故障识别准确率较高,出现错误识别的概率为0,峰值信噪比最高值达到了28 dB,说明所提方法的抗噪性能较好。Aiming at the problem of low accuracy of electrical equipment overheating fault identification in the existing methods,an electrical equipment overheating fault identification method based on three-phase self search and comparison is proposed.Preprocess the infrared image of electrical equipment,obtain the local maximum and minimum of the image through normalization processing,and convert the infrared image of electrical equipment into gray image.The edge features of the image of electrical equipment are obtained by laser sensor.The infrared image of electrical equipment is divided into parts with similar size and attitude by three-phase self search and comparison method,so as to judge whether there is abnormal temperature rise in the corresponding image block,judge whether there is thermal fault of electrical equipment,and realize the identification of overheating fault of electrical equipment.The experimental results show that the fault recognition accuracy of the proposed method is high,the probability of error recognition is 0,and the maximum peak signal-to-noise ratio reaches 28 dB,indicating that the anti noise performance of the proposed method is good.

关 键 词：激光传感器 电气设备 过热故障识别 图像增强 三相自搜寻比较 

分 类 号：TN219[电子电信—物理电子学]