Research on Intelligent Diagnosis Method of Power Grid Fault Components Based on Fault Fingerprint Technology

When the power grid fails, a large number of remote signaling alerts and dislacement information are uploaded to the dispatching terminal, which makes it very difficult for the dispatching personnel to make accurate judgements on the faulty equipment and faulty types in a short period of time. This paper prposes a method to form different fault fingerprints codes based on the remote signal displacement information of different faults and map the remote signal data to the fault diagnosis space. Using remote signal displacement data generated by different fault modes, Hoffield neural network is trained to correct remote signal misplacement or missing transmission data, and finally a fault diagnosis model is formed to realize the diagnosis of power grid fault elements in the fault diagnosis space. Through the test of the fault fingerprint data of the actual system, the validity of the Hoffield neural network information correction model and the fault diagnosis model for the diagnosis of grid fault components are verified.