A Multi-Scale Fusion Model for Grid Transformer Anomaly Detection

The power transformer is a crucial component of the power system, with its operational status directly impacting the stability and reliability of the power grid and forming an important part of safety management. However, traditional fault diagnosis methods struggle to meet the needs of real-time monitoring and face limitations in processing multi-scale data. To address these challenges, this paper proposes a transformer anomaly detection model based on a multi-scale sampling and inter-scale fusion mechanism. This model can effectively capture and fuse the multi-scale characteristics of voltage sequences. Through experimental verification using a real power grid data set, the accuracy and robustness of the anomaly detection model have been significantly enhanced. The results demonstrate that this model offers a novel technical approach for the intelligent maintenance of power systems. It helps to improve the accuracy of transformer monitoring and fault warning, and provides a new technical direction for the security management and big data analysis of smart grids.