Deep Independent Component Analysis-Long Short-Term Memory Model for Electromagnetic Interference in Substation Wireless Networks

Wireless communication in contemporary substations plays an essential role in monitoring, protection, and control applications. Nevertheless, high-voltage equipment-induced electromagnetic interference (EMI) and switching transients inflict substantial difficulty on the reliable transmission of data. Conventional interference cancellation methods depend heavily on filtering or threshold-based approaches, which cannot support flexibility and reliability within sophisticated substation atmospheres. This article suggests a hybrid approach of Independent Component Analysis (ICA) and Long Short-Term Memory (LSTM) networks for wireless communication interference identification and elimination in substations. ICA is used to separate statistically independent source signals, essentially filtering out interference from useful communication information, whereas the LSTM network utilizes temporal dependence to classify and forecast interference patterns with great accuracy. In comparison to conventional filtering and statistical models, the new ICA-LSTM model exhibits better performance with accuracy of 98.4% and F1-score of 97.8%, significantly outperforming baseline approaches. The adaptability of the model to changing interference conditions improves wireless communication reliability, rendering it extremely appropriate for real-time substation applications. This research confirms the efficacy of deep learning-based signal separation for reliable substation communication systems.