有限云模型和距离判别法在边坡稳定性评价中的应用

Application of finite cloud model and distance discrimination in slope stability evaluation

边坡稳定性分类是一个综合的不确定系统问题,针对指标分布和等级边界模糊的特点,提出一种以有限区间下的云模型为评价主体、距离判别法确定指标权重的稳定性分类的方法。给出有限云模型的相关概念及计算模型,再介绍具体的赋权的原理及流程。引用相关文献的评价指标体系,结合隶属函数计算出各等级的基本概率分配(BPA),并依据指标间距离确定指标权重;再依据改进的云模型计算各评价因子隶属于不同等级的云特征参数,并结合评价指标权重和云发生器,得到评价样本的综合确定度,实现边坡稳定性等级的划分,与其它理论方法对比分析。研究结果表明:该模型应用于边坡稳定性分级是有效可行的,能弥补传统云理论不能整体上描述有限区间下的指标间的确定与不确定性关系的缺陷,提高云模型应用的普适性。

Evaluation of slope stability is a comprehensive system problem with uncertainties, as characteristics index distributions and their classification boundaries are of fuzzy, this paper proposed a method to classify the stability of slope by using the cloud model with finite interval as the evaluation subject and the distance discriminant weighting. Calculation model and concept of finite cloud model and specific principles and processes of weighting were first discussed. And the evaluation index system of related literature was cited, the basic probability assignment of each index was calculated by the membership function and their weights were obtained based on indexes distance in the end. The cloud of each index on different stability levels was generated based on the cloud digital characters of evaluation indexes, and considering the measured evaluation index values and the corresponding weights, the comprehensive degrees of certainty were obtained. The slope stability level was identified by the maximum subordination principle. The results from case study and comparing with other theory methods show that the model proposed is feasible and effective, it can improve the shortcomings of inability to discuss quantitatively certainty and uncertainty relationships among indicators of interval distribution form from a unit way, so as to improve the universality for the cloud model application.