基于多源数据融合的GIS机械故障检测技术

Mechanical Fault Detection Technology of GIS Based on Multi⁃source Data Fusion

气体绝缘金属封闭开关设备(gas insulated switchgear,GIS)由于具有占地面积小、可靠性高等优势,在电力系统中被广泛使用。随着设备运行时间的增加,越来越多的故障出现在GIS内部。根据运行数据统计,机械故障所占的比例约占39.3%。部分机械故障还会进一步引起放电,导致电网运行事故。然而,目前并没有一种方法能够快速、有效地对故障进行准确与定位。因此,文中提出了一种基于多源数据融合的GIS机械故障检测技术。首先,利用声成像系统对GIS的运行噪声进行声源定位。其次,利用X射线成像系统,对该部位进行透视拍摄。最终,利用图像处理技术对GIS内部的焊接斑点进行自动识别,实现了对GIS内部故障的判别与定位。文中通过声源定位技术、X射线透视、图像识别技术等多元数据融合,提出了一套适用于GIS的机械缺陷诊断与定位方案,实现了GIS内部机械故障的准确定位,并有效提高了检测的速度与准确性。

Gas insulated metal enclosed switchgear(GIS)is widely used in power system due to its such advantages as small footprint and high reliability.With the increase of operation time of the equipment,more and more faults ap⁃pear inside GIS.According to the statistics of operation data,the proportion of mechanical failure is about 39.3%.Some mechanical faults will cause discharge further,leading to accident of power grid operation.However,there is no method at present which can quickly and effectively locate the fault.Therefore,a kind of mechanical fault detection technology of GIS based on multi⁃source data fusion is proposed in this paper.First,the acoustic imaging system is used to locate the noise source of the noise of GIS in operation.Then,the X⁃ray imaging system is used to take fluoros⁃copy of this part.Finally,the image processing technology is used to automatically identify the welding spots in the GIS so to achieves identification and location of internal fault in the GIS.In this paper,a set of mechanical defect di⁃agnosis and location scheme suitable for GIS is proposed through the multi⁃source data fusion such as sound source location technology,X⁃ray fluoroscopy and image recognition technology,which achieve accurate location of mechani⁃cal fault in the GIS and effectively improve the speed and accuracy of detection.