Research on Mining of Transmission Grid Assets of Heterogeneous System Based on Digital Twin

It is meaningful that power grid companies accurately assess the status of the power transformer and rationally arrange equipment maintenance, fault diagnosis, and replacement to enhance their management level and benefits and to ensure the safe and stable operation of entire power grid because the transformer is key equipment for power grid assets in power systems. However, it is difficult to accurately describe the attributes of assets by a single monitoring system. Therefore, it is necessary to organize all recording data related to power assets in different information systems and to dig out important attributes such as the value and life of assets, and then provide important support for the improvement of the management level of power assets. Currently, digital twins technology has received widespread attention in various regions to represent any object or its characteristics, but their application in power systems is still limited. Therefore, this paper proposes a method of transformer state assessment based on digital twins. Firstly, we combed the characteristics of multi-source heterogeneous systems, integrates and builds a complete set of asset attribute index systems, and then obtains sample data and labels from various systems. Secondly, we fully labeled the value of data and guided the actual state evaluation decision. Finally, we compared the sample data of the digital twin model with the labels and physical systems to improve the actual system's state evaluation capabilities.