| Popis: |
The proposed research aims to optically monitor the dry pollution level on transformer bushings and determine the possible leakage current should the dry polluted surface be critically wetted. The research involves the implementation of an image capturing system with appropriate image processing. Preliminary image capture of four artificial levels of salt deposit pollution: clean, light, medium and heavy is successfully achieved. The percentage level of surface pollution is found using image binary thresholding. For light pollution, the pollution surface coverage is found to be 8%, with each consecutive level increasing by 3%. These serve as part of the input to a neural network that will output the predicted dry pollution level and type in the image. The leakage current must be measured at various known pollution levels under wetted conditions. These values are stored as a reference that the neural network must then use to correlate the predicted dry pollution level to leakage current under wetted conditions. The standard methods used to classify pollution types and severity is presented. The dynamics governing bushing flashover under polluted conditions is discussed. The actual pollution level and type is quantified using Equivalent Salt Deposit Density (ESDD) and Non-Soluble Deposit Density (NSDD). Image segmentation and border extraction are illustrated to output four variables related to surface pollutants: area ratio, coverage, shape factor and eccentricity. The first two parameters are proposed as measures of surface pollution density, while the latter two may assist in pollution type identification. For more accurate pollution type identification, reflectance transformation imaging (RTI) is proposed. With a saliency mapping resolution of approximately 100 μm, the feature recognition between salt deposits and dust and dust deposits is more readily attained. |
