Abstract: This paper proposes an ontology-based dynamic risk control mechanism to address the challenges of dynamic risk management in tunnel construction, enhancing the automation and intelligence of risk identification and response. The study constructs a schema layer and an instance layer to explicitly define engineering entities and risk factors. By integrating Semantic Web Rule Language (SWRL) and Fault Tree Analysis (FTA), the framework implements rule-based reasoning for risk events, systematically embedding risk mechanism knowledge into the risk management process. The proposed mechanism is validated through a case study on water inrush risk control in the Second Jiaozhou Bay Subsea Tunnel. Leveraging ontology modeling, semantic reasoning, querying, and visualization technologies, the system achieves automated monitoring of tunnel risk data, along with real-time risk identification and analysis. Results demonstrate that this mechanism effectively identifies and analyzes dynamic risks in complex engineering environments, significantly improving the timeliness and accuracy of risk control. The key contribution of this study lies in the novel application of ontology and semantic web technologies to tunnel risk management, providing an innovative solution for dynamic risk control in large-scale underground projects.