Research on Real-Time Monitoring and Data Mining Methods for Subway Tunnel Construction Safety

In order to solve the problem of safety risk identification and real-time warning in complex environments during subway tunnel construction, this paper proposes a multi-source monitoring and data mining fusion method for construction sites. By constructing a GRU time series model based on feature evolution and a risk-driven mining process, dynamic identification and intelligent judgment of risk levels are achieved. The experiment is based on the 2024 subway tunnel flood intrusion monitoring data, and verifies the superiority of the model in terms of recognition accuracy, response delay control, and warning visualization.