基于综合决策云模型的围岩稳定性分级方法研究

Research on Classification Method of Surrounding Rock Stability Based on Comprehensive Decision Cloud Model

地下工程围岩稳定性分析关系到地下工程的安全,围岩的稳定与否直接决定着地下工程的成败,因此开展地下工程围岩稳定性评价具有十分重要的意义。基于云理论,选取岩石单轴饱和抗压强度、完整性系数、岩石基本质量指标、地下水影响修正系数和软弱结构面产状影响系数作为围岩稳定性分级指标,综合运用DEMATEL决策模型和熵权法获取上述指标体系的组合权重,并选取公路隧道工程20组实测数据作为学习样本,建立围岩稳定性分级的综合决策云模型,将其应用于小红石砬子玉石矿5组工程围岩的稳定性分级实例中。结果表明:综合决策云模型判别结果与实际情况相吻合,判别准确率可达80%,且优于K-最邻近算法和随机森林算法的判别结果,说明所建立的综合决策云模型在工程实际中具有一定的应用价值,为围岩稳定性分级提供了新思路。

The stability analysis of surrounding rock of underground engineering is related to the safety of underground engineering.The stability of surrounding rock directly determines the success or failure of the underground engineering.Therefore,it is of great significance to carry out the stability evaluation of surrounding rock of underground engineering.Based on cloud theory,the rock uniaxial saturation compressive strength,integrity coefficient,basic rock quality index,groundwater impact correction coefficient,and weak structure surface condition impact coefficient were selected as the surrounding rock stability classification indicators.The combination weight of the above index system was obtained by using DEMATEL decision-making model and entropy weight method,and then 20 sets of measured data of highway tunnel engineering were selected as learning samples to establish a comprehensive decision cloud model for surrounding rock stability classification.It is applied to an example of surrounding rock stability classification for 5 groups of Xiaohongshilazi jade mine.The results show that the discrimination result of the comprehensive decision cloud model is consistent with the actual situation,and the discrimination accuracy can reach 80%,which is better than the discrimination results of the K-nearest neighbor algorithm and the random forest algorithm,indicating that the comprehensive decision cloud model established in this paper has certain application value in engineering practice and can provide a new idea for stability classification of surrounding rocks.