A Cyber-Physical-Social System Framework for Power Outage Risk Factor Classification Using Deep Learning Method

In recent years, with the growing scale and technical complexity of power systems, power outages have become increasingly frequent, leading to serious socio-economic impacts. This highlights the critical need for automated analysis systems to mitigate outage risks. While power outage reports serve as a valuable resource for identifying risk factors, the heterogeneous and unstructured information of these reports poses challenges for accurate analysis. To address this issue, this study aims to explore the risk factors affecting the power system within the cyber-physical-social system (CPSS) framework and employs a deep learning method to classify the risk factors of power outage through related reports. Experimental results demonstrate that the proposed method achieves a precision of 0.9855, indicating high accuracy and effectiveness in identifying relevant risk factors. Furthermore, the system enhances outage management by utilizing cosine similarity to identify similar historical events and providing corresponding resolution references, thereby improving the overall performance of the analysis system.