Ground collapse is one of the most widespread and frequently occurring geological disasters all over the world. In China, more than 30 large and medium-sized cities, 100 prefecture-level cities and 420 counties are at high risk of ground collapse. Anthropogenic engineering activities, such as mining, tunneling, and underground space development, are the primary causes of urban ground collapse.
In the context of underground space development scenario, the inherent variability within composite strata for tunnelling is a major cause of ground collapse. In particular, the rock-soil interface (RSI) mixed ground, where residual or transported soils overlie bedrock, is one of the most encountered composite strata in shield tunnelling. The high spatial variability of the RSI interface introduces significant uncertainties in both subsurface stratigraphy and geotechnical properties. Inaccurate interpretation of these uncertainties during engineering geology investigations increases the risk of ground collapse.
Steady states and critical (or catastrophic) states alternate in most physical processes in the natural world. In the steady state phase, system parameters change slowly and are easy to observe, while in the critical state phase, the system changes abruptly and is difficult to observe. Catastrophe Theory is a mathematical theory based on differential manifold topology to study the change of dynamical systems, which considers the singularity as the critical point of the nonlinear system state change. Therefore, once the space-time coordinates of the geo-environmental catastrophe singularity can be predicted and identified, it is expected to realize accurate early warning of ground collapse and other geohazards.
Based on this insight, Associate Prof. Wei Zhang, in collaboration with Prof. Hong-Hu Zhu, developed a novel intelligent computational framework for the prediction and early warning of urban ground collapse in shield tunneling scenarios in RSI mixed ground. The framework employs the swarm intelligence algorithm optimized extended Kalmann filter to predict the amount of ground surface displacement at the future moments in real time after the multivariate fusion of the shield tunneling control parameters. Then, the gradient ratio criterion is used to identify whether the future ground surface displacement has entered the critical phase. If so, the moment is a singularity moment that triggers the early waning. The framework has been verified to successfully realize pre-disaster prediction and early warning of ground collapse induced by shield tunneling through a typical RIS mixed ground. It will help reduce the risk of ground collapse caused by material interface uncertainty and support the construction and development of resilient cities in China.
Fig. 1 Catastrophic differential manifold
Fig. 2 Singularity identification criterion
Fig. 3 RSI mixed stratigraphy transversed by the shield machine
Fig. 4 Early warning results of surface displacement prediction before and after ground collapse
This study was published in the journal of Engineering Geology under the title “Forecasting and early warning of shield tunnelling-induced ground collapse in rock-soil interface mixed ground using multivariate data fusion and Catastrophe Theory”. The full article can be accessed at https://doi.org/10.1016/j.enggeo.2024.107548. The authors acknowledge the financial support from the National R & D Program of China, the National Science Fund for Distinguished Young Scholars, the National Science Foundation of China, and the Jiangsu Provincial Key R & D Program.