Excavation Optimization and Stability Analysis for Large Underground Caverns Under High Geostress: A Case Study of the Chinese Laxiwa Project

In situ investigations and detailed laboratory tests indicated that the granite at the Laxiwa hydraulic station is a typical hard rock, with high compressive strength and elasto-brittle failure modes, such as spalling and slabbing, and that the underground caverns are prone to brittle failure. Thus, an intelligent optimization method for cavern excavation was developed to improve the underground engineering’s stability during its construction. This artificial intelligence method utilized the advantages of both the particle swarm optimization algorithm, which is capable of global optimization, and the support vector machine algorithm, which is capable of highly nonlinear mapping. The corresponding numerical analysis indicated that this optimization of excavation sequencing can considerably reduce both the total volume of the damage zone and the brittle failure of the surrounding rock. Furthermore, the measured deformations, the depth of the tested excavation damage zone, and the exposed in situ failures resulting from the applied excavation scheme were similar to the results predicted by the numerical simulation of the cavern excavation.