Abstract: Since the traditional classification of hydrophobicity grades of composite insulators mainly relies on power grid staff working at high altitudes,it’s difficult to ensure the quality of testing affected by factors such as environment and weather,resulting in low work efficiency.In this paper,a new classification method of composite insulator hydrophobicity based on KNN algorithm is proposed and the KNN algorithm is tested to select the most suitable parameters to classify the hydrophobicity grade of composite insulators. The experimental results show that,when K is equal to 8 and Manhattan distance is used,the accuracy of classifying the hydrophobicity grade of composite insulators is the highest,reaching 86.41%.