低压故障电弧图像特征提取及试验

Feature Extraction and Experiment of Low Voltage Fault Arc Images

为了研究负载功率、负载类型对电弧面积变化的影响,基于CCD高速摄像机,构建了高速图像采集系统。通过电弧图像的连续采集获取一个完整的燃弧过程图像。利用数字图像处理技术对图像进行去噪、图像增强、边缘检测、图像分割等处理,得到电弧图像面积。试验结果表明,在相同试验条件下,负载功率越大,电弧燃弧面积越大,燃弧时间越长,电弧危害程度越大;不同负载电弧燃弧面积和燃弧时间不同,为定量判定电弧危险程度提供依据。

The influence of arc area changes under different load power and load type was studied based on the CCD high speed camera. The whole images of the low voltage fault are, from burning to extinguish, were obtained. Digital image processing technique was used for noise suppression, image enhancement, border detection, image segmentation etc to count area of the arc images. The experimental results show that the bigger power of the same load is, the larger burning arc area is and the longer burning time is, that leads to the higher level of the arc danger. The arc area variation is also in connection with load type, which lays the foundation to quantitatively determine the arc dangerous level.