基于概念格粗糙集的采矿方法优选研究

为了正确选择矿山初选的采矿方法，提出基于概念格粗糙集的采矿方法评价体系。综合考虑影响采矿方法选择的众多因素后，对指标进行分层处理，利用改进的粗糙集建立采矿方法评价体系，生成最少决策规则集。属性约简是粗糙集中的核心问题，选择概念格作为约简工具，对条件属性进行约简。将模型用于15种采矿方法的优选，得到了最大可约简属性集，决策规则集的分类质量为100%。最后，将约简概念格与传统粗糙集中的分辨矩阵进行对比，结果表明：概念格在属性约简方面比分辨矩阵更有效，利用概念格的粗糙集构建采矿方法评价体系对矿山生产具有一定的理论指导意义。

Research on Mining Method Optimization Based on Concept Lattice Rough Set

While making rational use of mineral resources，it is necessary to make rational selection of mining methods so as to improve the utilization rate of mineral resources.The selection of mining methods has a decisive impact on the overall improvement of mine production capacity，safety and economic benefits，and also has direct effects on the degree of environmental damage caused by mining production.Therefore，it is particularly important to choose a scientific and reasonable mining method.In order to make a correct selection of the mining method for the primary section，to improve the efficiency of mining method optimization and perfect the process of mining method optimization，a mining method evaluation system based on concept lattice rough set is proposed.The evaluation system is aiming at the problems of strong subjective arbitrariness and insufficient analysis of index information in current mining method optimization，combining quantitative methods with qualitative ones，analyzing the relationship between evaluation index and mining method.Firstly，according to the idea of constructing evaluation system by analytic hierarchy process (AHP)，many influencing factors of the selection of mining method were comprehensively considered and sliced，so as to obtain the evaluation index system of mining methods with complete information.Then，the improved rough sets were applied to establish mining method evaluation system.As attribute reduction is the kernel contents of rough set，using concept lattice as reduction tool can get the maximal reductions，and minimum decision rule set was generated by using reduced evaluation index.Finally，the model is applied to the evaluation of 15 mining methods，and the classification quality of 15 mining methods is 100\mathrm{\%} according to the decision rule set.In order to verify the data processing ability of concept lattice for attribute reduction，the reduced concept lattice is compared with that of the traditional rough set.The results show that the concept lattice is more effective than the resolution matrix in attribute reduction.It can carry out deep data mining on evaluation indexes of mining methods.Based on the relationship between condition attributes and decision attribute，it can also reduce the indexes needed for evaluation of mining methods.As to the ability of rough set to deal with uncertainties，the minimum decision rule set generated by rough set can classify the pros and cons of alternative mining methods.The construction of mining method evaluation system by using the rough set based on concept lattice has a certain theoretical significance for mine production.