基于激光成像技术的电气设备故障自动分类研究

传统方法检测电气设备故障时,存在检测准确率低、耗时长的问题。为此提出基于激光成像技术的电气设备故障自动分类方法。对电气设备图像进行对比度拉伸处理,并采用滤波对其进行去噪处理,以提升对设备故障区域的检测能力,即故障区域在激光图像中亮度过大;利用脉冲耦合神经网络,将处理的激光图像过亮区域进行提取,获得电气设备故障区域,可通过激光图像中故障区域的亮度,实现电气设备故障类型的诊断和自动分类。真实验证明,研究方法的故障诊断性能较高,分类结果较为准确,且实现了非人工故障识别和分类,分类运行时间明显缩短。

Automatic classification of electrical equipment faults based on laser imaging technology

When the traditional method is used to detect the fault of electrical equipment,there are some problems,such as low detection accuracy and long time consuming.In this paper,an automatic fault classification method for electrical equipment based on laser imaging technology is proposed.The contrast stretching of the image of electrical equipment is carried out,and the filtering is used to denoise the image in order to improve the detection ability of fault area of the equipment,that is,the brightness of the fault area is too large in laser image.The pulse coupled neural network is used to extract the overlit region of the processed laser image to obtain the fault area of electrical equipment.The diagnosis and automatic classification of the fault types of the electrical equipment can be realized by the brightness of the fault area in the laser image.True experiment proves that the fault diagnosis performance of this method is high,the classification results are more accurate,the non-manual fault identification and classification are realized,and the classification running time is obviously shortened.