Study On Stability Analysis And Intelligent Optimization For Large Underground Caverns Under High Geostress Condition

It is very important to study on the stability and optimization for large underground caverns under high geostress condition, which is an unresolved key problem in current hydroelectric development. Aiming at unique of mechanics behavior of rockmass under high geostress and request for global and fast optimization of excavation schemes and support schemes for large caverns, using the large caverns of Laxiwa Hydropower Station as the applied engineering background, adopting new ideas, basing on numerical method combined with intelligent optimization tech, machine learning tech, a new method of stability analysis and intelligent optimization is proposed. The satisfied results about deformation and failure of rockmass and position of potential rockburst are obtained by the large-scale 3D elasto-plastic numerical simulation for excavation and support process of large caverns using a new constitutive model, a new method to identify parameters of constitutive model, a new rockburst index. Moreover, the stability of surrounding rock and various construction schemes are evaluated using new integrated index. The globally optimum excavation sequence and support parameters are achieved by a new intelligent global optimum method. Thus, the problem, which mentioned above, can be solved successfully by the new method. Several aspects have been studied including:1. Aiming at the problem, that is, traditional elsto-plastic constitutive model have not been successful in predicting the depth and extent of brittle failed of hard rocks under high geostress condition, a new hard rock constitutive model—cohesion weakening and frictional strengthening (CWFS) are adopted for solving this problem. In light of the fact that it is hard to determine parameters of the constitutive model, a new parameter intelligent identification method based on Particle Swarm Optimization (PSO) is proposed to solve the problem successfully.2. In response to the limitation of conventional index for stability and optimal design of underground caverns under high geostress condition and the drawbacks of conventional ERR rockburst index which based on linear elastic theory without brittle failure phenomenon, Local Energy Release Rate (LERR), a new rockburst index is proposed in the thesis. Using the new index, the intension of rockburst, the position and extent of brittle failure zone can be predicted quantitatively during rockmass excavation under high geostress condition, thus, more reasonable quantitative prediction for intensity and position of rockburst can be obtained.3. A new integrated stability evaluative index based on energy release rate, local energy release rate, failure zone, displacement, stress is proposed for stability analysis at underground caverns under high geostress condition in order to overcome the limitation of tradition stability evaluative index on predicting multi failure mode of surround rocks of caverns under high geostree synthetically.4. A new integrated optimization indexes including energy release rate, elastic release energy, plastic failed volume, displacement of key points of surrounding rocks and support cost is proposed to is proposed for stability analysis at large underground caverns under high geostress condition in order to achieve more scientific globally optimum result of optimization of caving and supporting scheme at large underground...