电流互感器红外故障热像图自动诊断方法

Automatic diagnosis method of infrared fault thermal image of current transformer

相对温差法是电力设备红外图像故障诊断的定性方法,但该方法易将电压致热型缺陷误判或遗漏。因此,文中提出用目标检测算法做图像分割处理,结合自适应滑动窗口技术提取温度特征和坐标信息判断电流、电压致热故障的诊断方法。其过程是,以目标检测算法的预测框坐标分割设备,设置自适应滑动窗口在分割后的图像中滑动,分别筛选出当前窗口中的最大温度值,将同位置中温度最小的一个作为设备的正常温度,其余的作为热点温度,根据窗口的位置判断热点温度的致热类型。以电流互感器为研究对象,根据致热类型相应的计算热点温度的相对温差或设备相同部位的温差来对设备进行诊断。实验表明,该方法最终能够对电流互感器的故障位置、故障类型和故障等级自动识别,能实现电流互感器故障的智能和高效诊断。

Relative temperature difference method is a qualitative method for infrared image fault diagnosis of power equipment,but it is easy to misjudge or omit the voltage induced heat defects.Therefore,this paper proposes an image segmentation method based on target detection algorithm,combined with adaptive sliding window technology to extract temperature characteristics and coordinate information to judge current and voltage thermal fault diagnosis method.The process is that,in the prediction of target detection algorithm coordinates integral equipment,set up the adaptive sliding window after the segmentation of the image,respectively,selected maximum temperature in the current window,with a minimum temperature in the location as the equipment of the normal temperature,the rest of the as a hot spot temperature,according to the position of the window of hot point temperature judgment to heat type.In this paper,the current transformer is taken as the research object,and the relative temperature difference of hot spot temperature or the temperature difference of the same part of the equipment is calculated according to the corresponding thermal type to diagnose the equipment.The experimental results show that this method can automatically identify the fault location,fault type and fault grade of CT,and realize intelligent and efficient fault diagnosis of CT.