Abstract:Using the Internet of Things technology, based on the cloud-side integration strategy, a system for monitoring and evaluating the overall status of the transformer is designed and developed.The system firstly senses the holographic state of the transformer to increase data diversity;secondly, it adopts edge computing to realize the sinking of computing power, improve the timeliness of data calculation in the station, reduce network transmission pressure, and transmit high-value data.Then a multi-source data fusion transformer state perception cloud model is build, and deep network plus machine learning and other technical means is applied to monitor and analyze the transformer state in real time, and enhance the robustness of the monitoring system.Finally, actual cases have verified the effectiveness of the system.