Surface Contamination Detection of Composite Insulators Based on Leakage Current

As an important part of the transmission line, the contamination accumulated on the surface of the insulator will cause pollution flashover accidents under the action of external conditions. [1], [2] Therefore, it is very important to predict the pollution state of the operating insulator in time to avoid the occurrence of flashover accidents. Firstly, the artificial contamination test was carried out on the ceramic plate brushed with RTV hydrophobic coating to obtain the leakage current data under different salt densities. Secondly, the characteristic quantities of leakage current are extracted: effective value, maximum value and standard deviation, which are used as input variables of BP neural network. Finally, the network model for judging the contamination state of insulators is tested. The results show that the three characteristic quantities can provide effective information for the detection of insulator pollution degree, and the network model for judging the pollution state of insulators can effectively predict the pollution state of insulators.