Power transformers are key components in the stable operation of electric power system. Reasonable arrangements of maintenance strategy according to transformers' working condition can decrease the loss lead by transformers' faults. The dissolved gas analysis (DGA) is the widest technique used in transformers' fault diagnosis and condition monitoring. Based on the conception of DGA technique, researchers and engineers have developed many methods and standard for faults diagnosis, such as IEC standard code, IEEE ratio, and Duval triangle method. They are practical and easy to use, but still face some problems. Many improvements for DGA have been carried out for improving the diagnostic accuracy. Artificial Intelligence (AI) method, statistics method or new diagnostic ways are the hot field for research. This paper has introduced the improved DGA methods of power transformer fault diagnosis in recent years and put forward some technical outlook of this field

No abstract

No abstract

Ultraviolet imaging technology is an effective method to detect the discharge of high-voltage electrical equipment. At present, the photon number is the main method to characterize the discharge severity of UV imaging. However, there is a complex nonlinear relationship between the parameter size and the gain and observation distance of the UV imager. The discharge was quantified. In order to quantify the analysis of discharge, based on the OpenCV image processing technology, the UV image was segmented by a combined threshold method, and the effective discharge region was extracted by the multi-region contour algorithm; Using the statistical information of foreground information pixel points, parameters such as area of discharge spot, perimeter, major axis minor axis, and spot profile are obtained. The method can quickly and efficiently judge the severity of UV fault images, save manpower and material resources, and have higher practical value.

Infrared image segmentation of power equipment is the basis for intelligent diagnosis of power equipment faults. In order to reduce the influence of non-uniform background on the infrared image segmentation of power equipment and improve the accuracy and efficiency of image segmentation, an improved Niblack image segmentation method based on bat algorithm is proposed. This method uses the variance between classes as the fitness function of the bat algorithm to automatically search for the optimal segmentation threshold of the non-overlapping rectangular neighborhood in the Niblack method, and uses it for binarization of the current neighborhood to extract target area from the infrared image. Experimental results show that compared with the traditional Otsu method, Niblack method and other algorithms, the segmentation algorithm reduces the ME by at least 34% to 84%. Compared with the BA+Otsu method, the average time consumption is reduced by 70%, effectively improving the accuracy and efficiency of the infrared image segmentation detection of the device.