In order to accurately detect power line faults and make proper classification, a power fault detection and classification system based on artificial intelligence is designed and implemented. Its core module works based on the support vector machine(SVM),a classical machine learning algorithm. To improve the accuracy of the model, its parameters are optimized by Harris hawks optimization(HHO). And the experimental results show that the SVM with the optimal parameters offers high accuracy for two open data sets. The system not only realizes the accurate power line fault detection and fault classification, but also streamlines and visualizes the upload, analysis and processing functions for fault data sets.