A BIM-based multi-model framework for advancing TBM performance − part 1: Real-time prediction of thrust force

Ping Xie, Ke Chen, Ziwei Yin, Yeting Zhu, Hanbin Luo, Qian-Bing Zhang

Research output: Contribution to journalArticleResearchpeer-review

4 Citations (Scopus)

Abstract

The prediction of Tunnel Boring Machine (TBM) performance typically requires analyzing a significant amount of multi-source heterogeneous data, including information from TBM machinery, geological conditions, logistical information, and operator-machine interactions. Throughout the excavation process, construction sources frequently undergo changes and often interact in complex ways, posing substantial challenges to accurate performance prediction. Building Information Modeling (BIM) has been extensively utilized for integrating data, managing construction, and facilitating dynamic interactions among multiple sources in construction projects. In response to these challenges and opportunities, we have developed a BIM-based multi-model framework designed to integrate and process multi-source heterogeneous data. This framework incorporates multiple Level of Detail (LoD) of TBM machinery, three-dimensional (3D) geological models and numerical analysis models. A non-intrusive TBM-GEO dynamic interaction method is proposed to efficiently analyze adjacent geological information in real time based on the TBM's spatial position. Additionally, a new BIM-to-Thrust analysis method is incorporated to achieve an automated workflow from BIM-based multi-model of shield tunneling to real-time thrust calculations. This method is applicable to both Mixshield and Earth Pressure Balance (EPB) shield projects. To validate this framework, two actual projects involving mixed ground conditions, confined water, and large curved tunnels are presented, and the results show that the efficiency and accuracy has been greatly enhanced in the analysis of shield thrust. The maximum average time for each analysis of these two case studies is approximately seven seconds. Additionally, comparisons with actual engineering data reveal that the average time for each simulation ring is significantly shorter than the average construction time per ring, confirming that the proposed analysis method meets the requirements for dynamic and real-time prediction of thrust force. We recommend promoting data sharing among different stakeholders and enhancing multidisciplinary collaboration in the prediction of TBM performance to further enhance these capabilities.

Original languageEnglish
Article number105856
Number of pages18
JournalTunnelling and Underground Space Technology
Volume151
DOIs
Publication statusPublished - Sept 2024

Keywords

  • BIM
  • Dynamic prediction
  • Numerical analysis
  • TBM performance
  • Thrust force

Cite this