Anti-Interference System for Power Sensor Signals under Artificial Intelligence

With the continuous improvement of China’s economic level, the scale of the power grid is also increasing, and it is developing towards high voltage, high current, and large capacity, which has led to a sharp increase in the probability of power sensor signal interference. This article first presents the general factors of signal interference in power sensors, and uses optoelectronic coupling isolation measures and twisted pair shielded wire long line transmission signals to design anti-interference signal transmission channels. This article regards the monitored equipment as a dynamic linear system and establishes an anti-interference system for power sensor signals using state space equations. In response to the weak temporal and spatial correlation of data collected by each node in power sensors, the GAN (Generative Adversarial Network) algorithm is used to filter a series of power sensor nodes. In the comparison of RMSE (Root Mean Square Error) and PSNR (Peak Signalto-Noise Ratio) of different algorithms, the RMSE (raw signal) of CNN (Convolutional Neural Network) is 0.23 and the RMSE (anti-interference signal) is 0.18. The PSNR (original signal) of GAN is 14.56, and the PSNR (anti-interference signal) is 17.89. The artificial intelligence based anti-interference system for power sensor signals helps to provide early warning for some anomalies, thereby improving the anti-interference ability of power sensor signals.