基于粗糙集综合权重法的公路隧道围岩等级预测方法研究

隧道围岩分类是考虑围岩所处的复杂地质水文环境及岩石自身条件等众多因素综合确定的。在收集已有隧道围岩分类数据基础之上，利用粗糙集理论对数据中的隐含知识进行挖掘，以期实现隧道围岩等级预判。结合样本数据，基于属性指标依赖关系及区分矩阵着重分析各个评价指标相对于最终围岩等级的权重，采用传统Pawlak权重计算和区分矩阵权重计算相结合的综合方法确定条件属性相对决策属性的综合权重，该方法克服了单纯依赖属性重要度条件下非核属性计算权重为零的不足。

Study on Prediction Methods of Highway Tunnel Surrounding Rock Class Based on Rough Set Comprehensive Weight Method

The tunnel surrounding rock classification is determined by considering the complex geologic hydrological environment of the surrounding rock, self conditions of the rock and other factors comprehensively. On the basis of collecting the classification data of the existing tunnel surrounding rock, explore the implicit knowledge contained in the data using the rough set theory so as to realize the class anticipation of the tunnel surrounding rock. Combining with the sample data, based on attribute index dependency relationship and discernible matrix, the weight of all the evaluation indexes relative to the final surrounding rock class is analyzed emphatically and the comprehensive weight of the condition attribute relative to decision attribute is determined using the comprehensive method combining traditional Pawlak weight calculation with discernible matrix weight calculation. This method overcomes the deficiency that the non-core attribute calculation weight is zero on the condition merely relying on the attribute significance.