Transformer Infrared Image Processing And Thermal Defect Area Identification

In order to meet the needs of the country's economic development,the power grid has basically achieved full coverage across the country.The "China Power Industry Annual Development Report 2020" pointed out that across the country,the loop length of 220 k V and above transmission lines is as high as 750,000 kilometers,and the capacity of 220 k V and above substation equipment is 4.3 billion k VA.There are a large number of substation equipment and transformers playing an important role in the operation of the power grid.Therefore,real.time monitoring of the operation status of the transformer is a necessary measure to ensure the normal operation of the power grid.When a transformer is in a fault state,it is generally accompanied by abnormal heating.Moreover,the oil.immersed transformer has a 6°C rule,that is,every time the temperature of the transformer increases by 6°C during operation,its service life is reduced by half.Therefore,measuring the temperature during operation of the transformer can accurately evaluate the state of the transformer.Compared with the thermometer temperature measurement method,analyzing the temperature of the transformer in operation through the infrared image of the transformer is more effective and real.time,and because it is a "non.contact" temperature measurement,it is also safe.In order to realize the identification of the thermal defect area of the transformer from the infrared image of the transformer,the main research contents of this paper are as follows:(1)Aiming at the Gaussian and salt and pepper noises formed in the imaging process of transformer infrared images,the method of wavelet transform combined with median filtering is used to denoise.First,transform and separate the noisy infrared image into a high.frequency domain with noise and a low.frequency domain with image information.Then,the high.frequency frequency domain is separated to reconstruct the image.Finally,median filtering is used to remove residual noise.By comparing the PSNR and EPI denoising effect indexes of the image after the denoising of each algorithm,the denoising effect of this method is better than that of the traditional infrared image denoising algorithm.(2)Aiming at the situation that the edge details of the infrared image are blurred,the target area and the background gray difference are small,an image enhancement method combining histogram equalization and bilateral filtering is adopted.First,perform histogram equalization processing on infrared images with concentrated gray values and low contrast.Then,the bilateral filter that has selected the best kernel value is used to smooth the image.The algorithm uses bilateral filtering to average the image slope while equalizing the gray level distribution of the image,maintaining the edge details of the transformer,and it also has the contrast evaluation index.Great improvement.(3)Aiming at the complicated and diverse backgrounds in transformer infrared images,an adaptive threshold method with improved SMD evaluation function is used to separate and extract transformer regions.First,the gray.scale variance product evaluation function is used to evaluate each gray.scale value as the threshold value of the infrared image after segmentation.Then,draw the relationship between the gray value and the evaluation value.Finally,the most stable area of the curve is solved by the gradient accumulation method,so as to select the optimal segmentation threshold.Through comparative experiments,the algorithm in this paper can more accurately extract the transformer area from the infrared images of many backgrounds than the commonly used infrared image segmentation algorithms,and the algorithm has a certain improvement in the objective evaluation index of the image segmentation effect.(4)Aiming at the influence of ambient temperature on the temperature of the transformer box,this article uses ANSYS software to perform finite element analysis of the transformer temperature field.First,according to the ambient temperature when the infrared image was taken,simulations were performed at ambient temperatures of0°C,16°C,and 22°C.Then,the maximum temperature and minimum temperature of the transformer under different ambient temperatures are obtained.Finally,according to the thermal fault temperature grade of power equipment divided in "Guidelines for the Application of Infrared Diagnosis Technology for Live Equipment",the fault temperature under the influence of ambient temperature is adjusted.