论文部分内容阅读
岩体质量分类是各类岩土工程实践问题的基础,由于岩体指标参数的模糊性和随机性,故岩体质量分类表现出不确定性,而现有分类模型则忽略了这种参数不确定性对分类结果造成的影响。考虑到分类指标参数的不确定性,在岩体质量分类问题中引入概率描述的可靠度分析方法,提出一种蒙特卡洛模拟(Monte Carlo simulation,简称MCS)和逼近理想解排序法(technique for order preference by similarity to ideal solution,简称TOPSIS)耦合模型。该模型由两部分组成:通过博弈论赋权方法获取分类体系中指标的权重,并结合TOPSIS模型来建立可靠度分析中的极限状态方程;基于MCS方法由TOPSIS模型构建的极限状态方程进行随机的不确定性分析,并通过概率函数给出最终的分类结果。通过25组测试样本对基于博弈论组合赋权的TOPSIS模型进行测试,结果显示该模型的误判率为0,表明MCS-TOPSIS模型中确定性分析的合理性。最后通过Matlab语言编制计算程序,分别由确定性分析和不确定性分析的角度来探讨水布垭地下洞室厂房围岩质量分类。研究结果表明:采用MCS-TOPSIS模型进行岩体质量分类是可行的,该模型具有较高的准确度且易于实现,具有一定的工程应用价值。该研究为岩体质量分类提供了一种新思路。
Rockmass quality classification is the basis of all kinds of geotechnical engineering practice problems. Due to the fuzziness and randomness of parameters of rock mass, the classification of rockmass quality shows uncertainty, while the existing classification model ignores this parameter The impact of certainty on the classification results. Considering the uncertainties of the parameters of the classification index, a reliability analysis method of probability description is introduced into the mass classification of rock masses. A Monte Carlo simulation (MCS) and the technique for approximation order preference by similarity to ideal solution, referred to as TOPSIS) coupling model. The model is composed of two parts: the weight of the index in the classification system is obtained by using the game theory weighting method, and the limit state equation in the reliability analysis is established by combining the TOPSIS model; the limit state equation constructed by the TOPSIS model based on the MCS method is randomly Uncertainty analysis, and through the probability function gives the final classification results. The test of TOPSIS model based on game theory combined weights shows that the false positive rate of this model is 0, which shows the rationality of deterministic analysis in MCS-TOPSIS model. Finally, the calculation program of the rock mass of the underground cavern house of Shuibuya Project is discussed by means of computational program of Matlab language, respectively from the point of deterministic analysis and uncertainty analysis. The results show that it is feasible to use the MCS-TOPSIS model to classify the mass of rock mass. The model has high accuracy and is easy to implement, and has certain engineering application value. The study provides a new idea for rock quality classification.