Distribution Network Fault Diagnosis Recorder

This paper describes a fault prediction and detection method of distribution network. It uses the ID3 algorithm in the decision tree to classify the grid parameters in the frequency domain after FFT transformation, and compares parameters with the built-in fault model in the terminal to find the corresponding fault type. The abnormal data needs to be recorded and stored at the same time and uploaded to the information center (cloud platform). The system can predict and monitor the abnormal parameters in a certain line timely, track the occurrence and evolution of the fault in time, and provide enough information to make decisions before the minor fault spreads to the regional grid.