Analysis Of Ultraviolet Radiation Characteristics Of Discharge On Contaminated Porcelain Insulators And Its Insulation Status Assessment

As the main insulation equipment for power transmission lines,porcelain insulators are affected by surface contamination,which can significantly reduce their insulation performance and increase the probability of pollution discharge and flashover.This seriously affects the safe and stable operation of the power grid,and timely assessment of their insulation status is of great significance.Ultraviolet(UV)imaging technology has many advantages in insulation discharge detection,such as strong anti-interference ability,long detection distance,non-contact detection,and accurate positioning of discharge location.It has been widely used in the detection of insulator discharge.Existing research usually uses photon counts or spot area as UV parameters to evaluate the insulation status of contaminated porcelain insulators.However,for different models of UV imaging equipment,the differences in optical system structure and signal processing algorithms can lead to significant differences in the numerical values of UV parameters measured by the instruments,resulting in a lack of consistency and comparability.Therefore,this paper focuses on studying the impact of factors such as temperature,humidity,air pressure,detection distance,and instrument gain on the process of generating and transmitting UV radiation in contaminated porcelain insulator discharge in the UV band.Starting from the integrated development of UV imaging equipment,consistency calibration of detection results,and improvement of UV image spot segmentation algorithms,a more applicable method for assessing the insulation status of contaminated porcelain insulators is proposed.The main research contents are as follows:A test platform for measuring discharge emission spectrum is built,and the solar-blind UV emission spectrum of glow-like discharge along the surface of porcelain insulators is obtained and its spectrum composition is determined.Based on the analysis of UV spectral characteristics,the relevant chemical reaction equations are determined.A two-dimensional simulation model of glow-like discharge along the surface of porcelain insulators is established using COMSOL Multiphysics software,and the radial distribution characteristics of surface electric field strength,current density,and electron density are obtained.By combining the optical parameters obtained through experiments with the particle parameters obtained through simulations,the correlation between temperature,humidity,and air pressure and optical and particle parameters is analyzed.The impact mechanism of temperature,humidity,and air pressure on the process of UV radiation in contaminated porcelain insulator discharge in the UV band is revealed.Based on the MODTRAN model,the impact characteristics of different factors such as temperature,humidity,air pressure,and detection distance on the atmospheric transmission process of solar-blind UV radiation are analyzed.A UV detection radiation transmission model is established,and a UV detection results consistency calibration method is proposed in combination with a radiation metrology consistency calibration platform.By using this method,a consistency calibration formula for the detection results of the UV imager is established,which converts the light spot area measured by the UV imager into the optical radiation intensity generated by the detection target.The stability of the results obtained by using the calibration formula is demonstrated through UV detection of corona discharge using a rod-plane model at different detection distances and instrument gains.Based on the emission spectrum characteristics of solar-blind UV discharge in the UV range and the UV detection radiation transmission model,two types of solar-blind UV imaging devices,including non-co-optical axis and co-optical axis configurations for UV and visible light,are integrated and developed.Firstly,a non-coaxial dual optical path structure is adopted,and the selection of the solar-blind UV filter,as well as the design of the optical system and the visible and UV image fusion system,are completed.A prototype of the non-co-optical axis UV imager is then fabricated.Then,a coaxial dual-path structure is designed,and a solar-blind induced transmission filter with higher half-width transmission is selected.Design parameters such as the aperture of the UV optical system,system focal length,and 80%diffraction spot radius are determined.After registering the visible and UV images using an affine transformation model and bilinear interpolation method,image fusion is performed using weighted averaging method.This resulted in the integration and development of a co-optical axis UV imaging system with higher imaging quality and smaller size.Calibration formulas for the two self-developed UV imagers are established,and the effectiveness of the UV detection results consistency calibration method is verified by using a deuterium lamp,the CoroCAM504 UV imager,and the two self-developed UV imagers.A 220kV transmission line porcelain insulator pollution discharge test is conducted and a non-white light spot mapping and segmentation algorithm for UV images with the addition of a Gaussian function preprocessing process.Based on discharge experiments,combined with surface discharge phenomena,leakage current signals of polluted porcelain insulators,and the UV image features obtained using a co-optical axis UV imager and a non-white light spot mapping segmentation algorithm with Gaussian function preprocessing,the insulation status of contaminated porcelain insulators is classified into four levels:"normal","fair","poor",and "very poor".The consistency calibration formula of the detection results using the co-axial UV imager is used to select three characteristic parameters based on UV radiation intensity.Taking environmental humidity,instrument gain and three characteristic parameters of light radiation intensity as input variables,and insulation status level as output variables,the intelligent optimization algorithm is used to optimize the parameters of support vector machine(SVM),and a suitable insulation status assessment model of contaminated porcelain insulators is established.