Voltage Transformer Fault Diagnosis Based on Improved ResNet50
摘要
In the modern power grid, the accuracy of power measurement and relay protection devices is critical for all parties involved in power supply, transmission, and utilization. The accuracy of the secondary current of the current transformer is of utmost importance in reflecting the primary current. To achieve this objective, we propose a fault diagnosis method for voltage transformers based on the ResNet50 network. This involves training the residual network with voltage fault signals to obtain an optimal fault diagnosis model with high recognition performance. By utilizing this model, we can accurately detect voltage transformer faults.