Large Language Model-Based Security Situation Awareness for Smart Grid: Framework and Approaches

The smart grid has integrated power generation, communication, and computing to deliver electricity efficiently, but its complexity and heterogeneity have exposed it to diverse security threats. Conventional security situation awareness approaches have struggled to manage the voluminous and heterogeneous data inherent to grid systems. This study introduces GridSense, a novel security situational awareness framework leveraging large language models to address the limitations of traditional methods and enhance predictive accuracy that has transformed electrical, meteorological, and social data into structured text prompts for large language models, leveraging their advanced reasoning capabilities. By employing prompt engineering and meta-learning techniques, GridSense has achieved robust performance. Experiments on real-world grid data in both full-shot and few-shot scenarios have demonstrated a multimodal load forecasting RMSE of 0.15 and an anomaly detection accuracy of 90.25% with an F1 score of 0.43 using combined data modalities. Even with limited training data, GridSense has maintained effective forecasting performance, showcasing its adaptability. These results underscore GridSense’s ability to integrate heterogeneous data effectively and highlight the transformative potential of large language models for smart grid security.