Substation secondary equipment fault detection based on the Llama2 model

Abstract

Abstract In the field of power systems, this paper proposes a novel method to detect and diagnose faults in secondary equipment in substations. Using the latest large-scale language model Llama2, a deep learning analysis of online monitoring information of substation secondary equipment was carried out to automatically detect and classify potential faults. The powerful semantic understanding and text analysis capabilities of the Llama2 model enable us to learn and identify failure modes from a large amount of historical data. When data annotation is limited, self-supervised learning technology is used to enable the model to learn and extract valuable features for fault detection from unlabeled data. Through a series of experiments and comparative analysis, our method not only improves the accuracy of fault detection but also reduces false alarms.