| Composite insulators are widely used in transmission lines due to their excellent anti pollution flashover performance.Their excellent anti pollution flashover performance is attributed to their surface hydrophobicity and hydrophobicity migration.However,as the service life increases,its pollution flashover resistance will gradually decrease,which is intuitively manifested as a decrease in the hydrophobicity of the surface of the composite insulator umbrella skirt.Therefore,it is necessary to test the hydrophobicity of the composite insulator.The traditional methods for detecting the hydrophobicity of composite insulators include the water spray classification method and the contact angle method.The former has convenient detection conditions but has significant subjective human factors,while the latter has more accurate detection results but limited experimental conditions.Based on the above analysis,this article proposes a method for the hydrophobicity of composite insulators based on image processing technology,which can quantitatively describe the surface hydrophobicity of composite insulators under on-site operation conditions.Firstly,the spray water images of composite insulators with 7 different hydrophobic levels obtained from the experiment are grayed,histogram equalization,median filtering and other image enhancement processing.The maximum inter class variance threshold segmentation algorithm is selected to obtain the binary image of the droplet.A description method of the surface hydrophobicity of composite insulators is proposed with the droplet area,perimeter,form factor and other characteristics,and the machine learning model based on the decision tree algorithm is used,Effectively identified 7 composite insulator water spray images with different hydrophobicity levels.Then,the static contact angles of liquid droplets on the surface of composite insulators with different hydrophobicity were obtained by using the contact angle tester.At the same time,these liquid droplets were photographed overhead,and the side and overhead images of the static contact angles of liquid droplets from 30°to 130° were obtained,The correlation curve between the static contact angle of droplets and the feature quantity of droplet aerial images was obtained,and a prediction method based on droplet image features was proposed to distinguish the distribution of static contact angle on the surface of composite insulators.Finally,the above discrimination method was used to comprehensively evaluate the droplet state of the insulator water spray image.The distribution of static contact angle on the surface of the composite insulator was calculated based on the area of water droplets on the surface,and compared with existing HC levels to achieve quantitative description of HC classification;Finally,the concept of weighted contact angle was proposed by integrating the area of water droplets/film on the surface of insulators and the static contact angle,which can quantitatively distinguish the hydrophobicity of composite insulators.The research results can provide a basis for quantifying the hydrophobicity of composite insulators. |
