An Anomaly Monitoring and Early Warning Method for Power Grid Microservice Network Based on Log Visualisation and Analysis

With the increasing complexity of power grid microservice networks, monitoring and detecting anomalies in real-time has become a significant challenge. Traditional methods struggle to capture the sequential and contextual relationships in log data. To address this, we propose a novel anomaly monitoring and early warning method based on the Bi-LSTM-ATT architecture, which integrates bidirectional long short-term memory (Bi-LSTM) networks and an attention mechanism. This model effectively captures both forward and backward dependencies in log sequences while focusing on critical features related to anomalies. The proposed method was tested using real-world log data from power grid microservices, and the experimental results show that it significantly outperforms traditional approaches such as PCA, Invariant Mining (IM), and N-gram in terms of precision, recall, and F1-score. The Bi-LSTM-ATT model provides a robust and accurate approach for real-time anomaly detection, contributing to enhanced operational stability and reliability in power grid systems.