基于红外热成像技术的配电柜故障监测与诊断

发布时间：2018-05-30 09:40

本文选题：红外热成像 + 图像处理　；参考：《安徽理工大学》2017年硕士论文

【摘要】：红外热成像技术是一种能够实现对目标物体实时观测,同时是根据目标设备的红外辐射探测来捕获目标的图像技术。本文提出一种电气设备红外图像自动分析和处理方法,对配电柜中的特定电气设备的过热现象现象进行温度监测和故障判别,实现了故障诊断领域中追求的高效与精准。本文对图像进行预处理主要使用了空域增强算法当中的灰度增强算法以及中值滤波,然后运用运用K-means聚类算法,从周围复杂环境中分割和提取图像边缘特征。本文红外识别主要针对两个部分、三个对象展开,分别是互感器和三相电力设备,三相电力设备中又选取了套管和避雷器来开展研究。对于互感器而言,通过比较多种算子的优缺点,最终确定利用Canny算子提取图像特征。归一化相关系数法将目标与模板图像的特征匹配后经过K-means聚类算法确定了边缘纹理特征,对离群点直线拟合最终定位故障位置。三相电力设备识别采用数学形态学的方法处理图像后,依照面积和重心坐标信息来判断三相电力设备各自的区域,以便明显划分界限进行识别。采用温度识别方法的目的是通过比色程序对被检测电气设备进行温度分析,并测算出各个温度区域的面积。相对温差法的温度识别方法既方便准确可计算,又能科学地读出故障的严重程度。相位拓扑矩阵能能对故障自动定位,改进了缺陷。根据国家规定的标准,应用不同的算法识别并诊断出配电柜中设备是否存在故障、故障类型和严重程度,形成了套管、互感器、避雷器的故障诊断完整流程,以此来实现对配电柜中故障设备的监测和诊断。
[Abstract]:Infrared thermal imaging is a kind of image technology which can realize the real-time observation of the target object and capture the target according to the infrared radiation detection of the target equipment. In this paper, an automatic analysis and processing method of infrared image of electrical equipment is put forward. The overheating phenomenon of special electrical equipment in distribution cabinet is monitored by temperature and fault discrimination is carried out, which realizes the high efficiency and precision in the field of fault diagnosis. In this paper, image preprocessing mainly uses gray enhancement algorithm and median filter in spatial domain enhancement algorithm, and then uses K-means clustering algorithm to segment and extract edge features from the surrounding complex environment. This paper focuses on two parts, three objects, mutual inductor and three-phase power equipment, and the casing and arrester are selected in the three-phase power equipment to carry out the research. For transformers, by comparing the advantages and disadvantages of many operators, the Canny operator is used to extract image features. The normalized correlation coefficient method matches the feature of the target with the template image and then uses the K-means clustering algorithm to determine the edge texture feature and to fit the outlier straight line to locate the fault location. After three phase power equipment recognition is processed by mathematical morphology, the area of three phase power equipment is judged according to the information of area and center of gravity coordinate, so that the boundary can be clearly divided. The purpose of the temperature recognition method is to analyze the temperature of the tested electrical equipment by colorimetric program and to calculate the area of each temperature region. The temperature recognition method of relative temperature difference method is convenient and accurate to calculate, and can scientifically read out the severity of fault. The phase topology matrix can locate the fault automatically and improve the defect. According to the standards prescribed by the state, different algorithms are used to identify and diagnose the fault, type and severity of the equipment in the distribution cabinet, thus forming a complete process of fault diagnosis of the casing, transformer and arrester. In order to realize the monitoring and diagnosis of the fault equipment in the distribution cabinet.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM642;TN219

【相似文献】