BUSINESS SCOPE
Key Technologies for Automated Monitoring and Early Warning of Structural Safety in Highway Tunnels
Project Overview:
This project adopts a combined approach that integrates “geological analysis, comprehensive geophysical exploration, and advanced drilling,” “the combination of short- and long-range forecasting methods,” and “surface-based and in-tunnel detection techniques.” Leveraging cutting-edge geophysical technologies and equipment with strong detection capabilities and high precision—such as transient electromagnetic instruments, the GeodeEM3D three-dimensional networked distributed electromagnetic exploration system, and the TSP303—we conduct specialized, detailed, and integrated geological forecasts for high-risk sections during tunnel construction, while mutually validating these forecasts. Building on the traditional single-method geological radar approach, this method further enhances the accuracy and level of detail in geological forecasting, thereby establishing a comprehensive highway tunnel forecasting system. This system can reduce the likelihood of geological hazards during tunnel construction, ensure construction safety, and provide robust technical support for “dynamic design adjustments, flexible construction practices, and the rational optimization of construction schedules and investment.”
The highway tunnel forecasting system, which employs a two-stage algorithm, can effectively achieve feature recognition of the surrounding rock at the tunnel face. Building on this foundation, we have improved the Mask R-CNN instance segmentation algorithm and developed an intelligent method for identifying and classifying the surrounding rock at the tunnel face.
Project Highlights:
1. The project team has integrated technologies such as the Internet of Things, cloud computing, sensors, and wireless communication to develop an intelligent, secure, and efficient automated monitoring and early-warning platform for highway tunnel structural safety. This platform provides effective data support for advanced geological forecasting, construction phase monitoring, and maintenance-phase management of tunnels.
2. Building on the two-stage algorithm, we improved the Mask R-CNN instance segmentation algorithm and proposed an intelligent method for identifying and classifying surrounding rock in tunnel face areas. We also developed a highway tunnel forecasting system to ensure tunnel construction safety.
3. Utilizing neural network algorithms to predict tunnel vault deformation, we have developed a neural network regression prediction model and established an intelligent management system for highway tunnel construction quality information. This system enables visualization of tunnel construction data and provides intelligent early warnings, conducts safety assessments of the construction process, and helps reduce safety hazards.
4. We have developed an automated monitoring system for the operational phase of highway tunnels, which provides a comprehensive overview of the tunnel environment, enables rapid collection of monitoring data, and facilitates real-time analysis, processing, and transmission, thereby achieving precise monitoring of the tunnel’s condition. The system automatically identifies abnormal conditions in the tunnel and issues tiered alerts, ensuring tunnel safety.
Honored Achievements:
Key Promoted Achievements of the “Four New Technologies” for Highway Tunnels by the China Highway Society in 2024.
Address: No. 100, Yangguan Avenue, National High-tech Industrial Development Zone, Guiyang City, Guizhou Province
mobile client
