Online Performance Evaluation Method for Transmission Line State Monitoring Sensors

With the increasing demand for electricity, the traditional power grid can no longer meet people's daily power needs, and the smart grid has become the development direction of the Chinese power grid. As the “eyes” of online monitoring in the smart grid, sensors play an increasingly important role. A large number of field practices have found that the abnormal performance or failure of sensors in transmission line state monitoring devices is one of the important factors affecting the effectiveness of online monitoring applications. The existing transmission line online monitoring systems mainly evaluate and analyze various monitoring data and the equipment health status reflected by them, but there are few studies on directly evaluating the performance and diagnosing faults of the sensor body. This paper proposes a performance evaluation index for transmission line state monitoring sensors based on measurement uncertainty, and then uses an LSTM neural network to predict the sensor output and calculates the measurement uncertainty based on the predicted value and the actual measured value, and proposes a performance evaluation method for transmission line state monitoring sensors. The results show that the method based on LSTM neural network in this paper can conduct real-time performance evaluation of transmission line online monitoring sensors, which is of great significance for improving the reliability of transmission line online monitoring devices.