Research On Key Technology Of SF₆ Gas Leakage Infrared Imaging Detection In Substation

Sulfur hexafluoride（SF₆）gas is mainly used for equipment insulation,isolation,and arc extinguishing,etc.,which plays an important role in the stable operation of power equipment.However,with the aging and wear of power equipment,SF₆ gas will also leak.Once SF₆ gas leakage occurs,it will affect the insulation strength of high-voltage electrical equipment,and even cause safety accidents.Therefore,it is very important to study an efficient and convenient SF₆ gas leakage detection method.With the maturity of infrared detection technology,passive infrared detection technology is widely used in power inspection,infrared monitoring,military warning and so on.Compared with the traditional contact physical detection method,the passive infrared imaging based gas leakage detection has the advantages of no power outage,reusable and high detection security,which is the development direction of intelligent detection of equipment in the future.However,when the gas（SF₆）leakage is small,there are usually the following problems:the surface and edge of the target object become blurred,the image details are lost,and the infrared image contrast is low;Limited by the hardware cost of infrared imager,the resolution of infrared image obtained by conventional infrared imager is low,usually 320×240;When the SF₆ gas leakage is small,the complex background and noise have a great impact on the target detection,leading to false detection.In view of the above problems of SF₆ infrared detection in substation,the main research work of this paper is as follows:（1）In this paper,the noise distribution characteristics of SF₆ gas infrared imaging are analyzed.An improved dynamic inter frame time domain filter is proposed to suppress the Gaussian like noise in the infrared image,and a local enhancement algorithm is used to enhance the details of the image to enhance the obvious degree of SF₆ gas leakage characteristics.（2）An on-line detection method of SF₆ gas small leakage based on Gaussian mixture model is proposed.Firstly,the random noise component in the image is suppressed by the improved dynamic inter frame time domain filter as the preprocessing stage,and then the dark details of the infrared image are enhanced by CLAHE to improve the SF₆ local contrast,Finally,the improved Gaussian mixture background model is used to segment the SF₆ leakage area with motion features adaptively and mark the leakage area.Finally,the automatic detection and location of small SF₆ leakage is realized.（3）The calibration and estimation model of SF₆ gas leakage is constructed.This paper analyzes the image features of SF₆ gas leakage,including boundary texture,motion diffusion characteristics,etc.,integrates multiple features to realize the comprehensive judgment of suspected area,estimates the size of SF₆leakage,and provides four levels of leakage standards based on the existing experimental data,which provides visual reference for equipment repair and maintenance.（4）The software system is designed and developed,and the application test is carried out.The algorithm is deployed and applied,and the host computer software based on Qt graphic environment is developed.The image algorithm is combined with the host computer software,and the adjustable parameters are combined with the visual control.The application efficiency of image processing algorithm in different scenes is improved,the time of parameter adjustment in different scenes is shortened,and the convenient operation and processing platform is provided for different people.