绝缘子是变电站的重要供电设备，绝缘子劣化会导致绝缘子闪络或掉串，对电力系统的安全稳定运行具有严重威胁。红外检测技术在瓷质绝缘子方面的应用是一种可靠、高效的检测手段，但使用目前的红外检测标准会造成多数劣化绝缘子的漏判。针对这种情况，笔者通过一起劣化绝缘子案例，基于劣化绝缘子发热原理，设计图像分割算法求出钢帽以及盘面区域，然后提取盘面以及钢帽温度特征，分析钢帽与盘面的发热规律。针对绝缘子钢帽与盘面的发热规律，提出相应的诊断方法及其预防措施，进一步提高劣化绝缘子红外检测的准确性。

Abstract：Insulator is major power supply equipment of substation. Degradated insulator can lead to insu- lator flashover or bunch-drop, and it is a serious threat to safe and stable operation of the power system. In the application of porcelain insulator detection, infrared detection technology is a reliable and efficient de- tection means. Application of current infrared detection standards will cause the leakage of most degradated insulators. In this paper, an infrared detection case is analyzed by designing image segmentation algorithm to calculate the steel cap and disk areas and extract the temp characteristic to analyze the law of fever based on heating mechanism of degradated insulator. In view of the insulator steel cap and disk fever regularity, and the corresponding diagnostic methods and preventive measures are put forward to improve the accuracy of the infrared detection for degradated insulators.