论文部分内容阅读
离散裂隙随机网络模型中结构面的分布特征仅代表其统计规律,并非地质成因、结构模式、构造形迹的客观反映,以致后续力学计算及渗流模拟结果可信度偏低。针对国家高放废物处置库甘肃北山预选区出露良好的花岗岩体,基于实测结构面统计规律、内在成生关系及水力联系,在前期纯随机模型关键部位修正特征参数,并融入人工辨识的确定结构模式而构建“随机-确定”耦合模型。图形和渗流模拟两种定量检验结果表明:耦合模型结构面数量更接近客观实际,模型准确度提升约48.8%;耦合模型渗流路径与流量较之前更显客观真实,模拟结果与试验数据的接近程度比随机模型大约1/3。另外,不同孔位渗流结果显示:相比于随机结构面,确定性结构面对区域渗流控制作用更加明显,在渗流模拟中扮演更为重要角色。此耦合模型有益于拓宽结构面网络模拟的发展方向。
The distribution characteristics of structural plane in the discrete stochastic network model represent only the statistical laws, not the objective reflection of geologic causes, structural modes and structural features, so that the follow-up mechanical calculation and seepage simulation results have low confidence. Aiming at the granite body exposed from the national high level radioactive waste disposal reserve in the Beishan preselected area of Gansu Province, based on the statistical laws of the structure plane, the inherent relationship between birth and birth and the hydraulic relationship, the key parameters of the purely stochastic model were modified and integrated into the artificial identification Structure model to build a “stochastic - deterministic” coupling model. The results of the two quantitative tests show that the number of structural planes of the coupled model is closer to the objective reality and the accuracy of the model is improved by about 48.8%. The coupled model seepage flow and flow are more objective and true than before, and the closeness between the simulation results and the experimental data About a third of the random model. In addition, the results of seepage at different positions show that the deterministic structure is more obvious than the stochastic one, which plays a more important role in seepage simulation. This coupling model is beneficial to broaden the development direction of structural plane network simulation.