一种电力设备红外热图像等温线绘制算法

摘要

Intelligent surveillance of power equipment is key technology in power distribution system. In order to predict and diagnose rapidly the potential failure point of power equipment, the traditional testing way is changing. Based on the characteristics of infrared thermal image, this paper describes a algorithm which uses the improved FCM clustering algorithm to separate the background and objectives of the images. Meanwhile, it detects automatically the position of the highest and lowest temperature of power equipment, then computes amount of isotherms and temperature interval to classify target power equipment. At last it draws the isotherms of the target area. The experimental results show that the method achieved a good practical effect. It can help technicians to predict possible failure points of power equipment and be worthy to applied in practice.%电力设备运行状态的智能监测是电网可靠运行的关键技术.为了快速预测诊断出电力设备潜在的故障位置,传统电力设备检测方式正在改进.针对红外热图像的信息特征,应用改进的FCM聚类方法将电力设备红外热图像的背景和目标分开.自动确定电力设备红外热图像的温度最高点和最低点,根据等温线数量和温度间隔,对感兴趣目标的温度进行分级,绘制出电力设备红外热图像中目标区域的等温线.实验结果表明,该研究方法达到了很好的实际效果,可帮助技术人员预测出多处可能发生故障的电力设备,值得实际推广应用.