Application of numerical analysis principles and key technology for high fidelity simulation to 3-D physical model tests for underground caverns
Zhongkui Li,Hui Liu,Rong Dai,Xia Su
DOI: https://doi.org/10.1016/j.tust.2005.01.004
IF: 6.9
2005-01-01
Tunnelling and Underground Space Technology
Abstract:Numerical simulation and model testing are the main research methods of solving geo-technical and underground problems. In the 1980s, physical simulation (model testing) was one of the most popular methods to solve these types of problems, but since then, with the rapid development of IT technology, numerical simulations have rapidly become dominant. However, the complexity and uncertainty of geological conditions and geo-technical engineering problems represent a great challenge to the numerical method, leading some researchers to question how far numerical simulation can go. In this case, the physical simulations have begun to reappear in research and the use of physical models has seen valuable progress. In the past, problems have been solved either with physical simulation or numerical simulation, or just by comparing the results of these two methods simply. But very little attention has been paid to the question of how to combine the advantages of both and improve them. In the 3-D geo-mechanical model of the underground caverns of the Xiluodu hydropower station, the advantages of the two methods were combined and some of the primary principles of the numerical method were combined with the physical model process. Considering that the boundary condition has a great effect on the stress distribution, a larger test model was set up. The numerical discrete principle was also been used to simulate an initial geo-stress field. The basic principle of cavern excavation according to requires that loading comes first, followed by excavation. According to this idea, we developed a new technique, in which a high fidelity simulation of the excavation process and a very difficult cavern excavation were successfully achieved for the first time. The new method produced test results more satisfactory, and this simulation represents an obvious improvement of existing techniques. We also conducted a numerical simulation with almost the same conditions, and compared the results of the model test and the numerical simulation.