Digital Twin Technology for State Monitoring of Power Transformers Based on Supercomputing Platforms

Monitoring the status of power transformers contributes to timely maintenance and reduces accidents. By connecting actual transformer sensor data to digital twin models and leveraging the high-performance computing capabilities of supercomputing platforms, accurate monitoring, fault prediction, and optimization of maintenance decisions for power transformer operation can be achieved. Real-time monitoring methods involve collecting real-time transformer parameter data and processing it in the digital twin model, enabling the temporal detection of faults and providing valuable solutions. Through continuous state prediction, digital twin technology can proactively identify potential failure risks and issue warning signals. Lastly, by analyzing the extensive statistical results generated by the digital twin model, more reasonable and precise maintenance plans can be formulated, providing crucial support for the safe operation and equipment maintenance of power systems. This study was conducted in a local area power network and demonstrated significant efficiency improvements and reliable risk warnings in both power transformer status monitoring and prediction.