Research on Transformer Fault Diagnosis Method Based on Digital Twin Technology

Abstract

Real-time fault diagnosis of transformer operation is an important approach to protect the safe and stable operation of the power system. The traditional diagnosis approach has some problems such as low data quality, low fault diagnosis efficiency, and difficult diagnosis model construction, so a transformer fault diagnosis model based on a digital twin is proposed. Firstly, the physical entity of the transformer is combined with the virtual model, and the digital twin model is modified according to the condition monitoring data obtained from the sensor and the simulation data of the twin model, and the characteristic parameters are extracted. Then the BP neural network algorithm is used to diagnose the types of faults and analyze the possible causes of faults, which can help reduce the maintenance cost and cycle of transformers, improve the efficiency of fault diagnosis, and make the transformers operate safely and reliably.