隧道围岩分级判别的未确知均值聚类模型

Application of Uncertainty Average Clustering Measurement Model to Classification of Tunnel Surrounding Rock

基于未确知测度理论,建立隧道围岩分级的未确知均值聚类分析模型。针对隧道围岩分级判别中等级评价中许多不确定性影响因素,选用岩石等级、风化程度、岩体弹性纵波波速、岩体结构、地质构造影响程度、节理裂隙发育程度和地下水情况等7个指标作为隧道围岩分级的判别因子;以20组隧道围岩实测数据作为训练样本,建立各评价因子的未确知测度函数,用各分类样本平均值表示其分类中心;根据信息熵理论计算各评价因子的权重,依照置信度识别准则进行等级判定;用建立的模型对20组实测数据逐一进行回检,正确率为100%。将建立的模型对待分类的10个样本进行测试,并与实际结果进行比较。研究结果表明:该模型判别预测结果与实际结果吻合,比较客观地反映了隧道围岩分级的复杂状况;且方法科学合理,意义明确,为隧道围岩分级判别提供了一种新思路。

Based on the uncertainty measurement theory, a uncertainty average clustering measurement model for tunnel surrounding rock classification was established. Due to the uncertain factors in judging the engineering quality of rock masses, seven indexes, i. e. , the rock grade, rock weathering degree, rock mass structure, elasticity longitudinal-wave velocity of rock mass, influence degree of geological structure, development of joint fissure and ground water regime, were used as the discriminating factors, the indexes functions of unascertained measurement of 20 sets of rock samples were established, and the centre of the classification was indicated by using the average of classification of samples. The weight of indexes was calculated by entropy weight theory, and a prediction for the classification of residual tunnel surrounding rock was carried out using the rules of credible recognition. Each of the 20 sets of tunnel surrounding rockmass samples was tested according to the model, and the correctness rate is 100%. The other 10 sets of tunnel surrounding rock samples were predicted by using this model. The results show that the uncertainty measurement model classification agrees well with the actual measured ones. Therefore, it shows that the uncertaity measurement model is effective, available and can be applied to classification of tunnel surrounding rock in underground engineering.