Research On Application Of Infrared Thermal Imaging Technology In Fault Diagnosis Of Substation Equipment

The State Grid Corporation of China proposed a new strategy of "three types and two networks" in 2019.Among them,the "two networks" are the strong smart grid and the ubiquitous power Internet of Things.Real-time monitoring of power equipment and timely detection and elimination of potential safety hazards are important links.Infrared thermal imaging technology is an effective method to find thermal defects in substation equipment and ensure the safe and stable operation of the equipment.It does not directly contact the substation equipment.According to the temperature distribution on the surface of the equipment itself,it detects areas with excessive temperature and does not stop the operation of the equipment.Next,a device failure is detected.Therefore,this article will be based on the technical characteristics of infrared thermal imaging technology in the fault diagnosis of substation equipment.Through the image preprocessing,image segmentation and equipment identification of the collected infrared images,the relative temperature difference method will be used to realize the infrared thermal imaging technology.Fault diagnosis of substation equipment.First of all,the images currently used in engineering are noisy and have low contrast.Aiming at the problem of image noise,a power equipment denoising algorithm based on adaptive median filtering is used.It can effectively alleviate the contradiction between noise reduction and image detail preservation.Simulation experiment results show that this algorithm has a significant denoising effect.Aiming at the problem of low image contrast,a method for infrared image enhancement of power equipment based on NSCT and improved Pal＿King algorithm is proposed.Use histogram two-way equalization to process infrared temperature measurement images,improve image clarity,and modify the coefficients of the sub-bands generated by NSCT transformation,reconstruct the image through NSCT inverse transformation,and apply new membership functions and improved blur Contrast,select the appropriate nonlinear transformation function to improve the original Pal＿King algorithm to improve the contrast of the image.The simulation experiment results show that the average gradient value,peak signal-to-noise ratio value and entropy value of the output image of this method are maintained above 5.9,25.4 and 6.8 respectively,which improves the contrast of infrared images of substation equipment.In order to effectively segment the infrared image of substation equipment,an infrared image segmentation method of substation equipment based on OTSU and improved area growth is proposed.The ROI of the image is separated by the improved OTSU algorithm,and the traditional region growing method is improved.The Sobel gradient operator is used as the direction limiting condition to quickly and accurately determine the direction.The simulation experiment results show that the method can clearly show the contour characteristics of the infrared thermal imaging device,the segmentation quality is better,the error segmentation area ratio is reduced to 0.348%,and the efficiency and accuracy of subsequent fault recognition are improved.Finally,intelligent identification and fault diagnosis of the infrared image of the equipment are carried out,and the method of infrared image identification of power equipment is studied.Use the ASIFT algorithm to match the image to be recognized with the picture of the substation equipment in the standard library and identify the type of power equipment.After obtaining the corresponding temperature information,determine and calculate the relative temperature rise according to the relative temperature difference method,and according to the determined temperature threshold Carry out fault diagnosis,determine whether the fault has occurred and the level of the fault,and give corresponding treatment opinions,so as to complete the fault diagnosis of the substation equipment.