“AHP+熵权法”的CW-TOPSIS煤矿内因火灾评价模型

CW-TOPSIS mine internal caused fire evaluation model of “AHP + entropy weight method”

煤矿内因火灾评价指标具有灰度、未知性等不确定性,难以确定合理的指标权重,导致评价不准确。针对此问题,构建了以煤层地质赋存、开拓开采技术、煤层自燃倾向性、通风条件及采空区管理4个因素为准则层,煤层厚度、煤层倾角、煤层岩性等24种因素为指标层的煤矿内因火灾评价体系;提出了一种组合赋权的方法,即AHP(层次分析法)的主观赋权与熵权法的客观赋权相结合,引入拉格朗日函数,构建优化决策模型,通过欧氏距离函数,保持了主、客观权重与偏好系数的一致性,得到指标组合权重,评价指标的主次关系,并运用TOPSIS(逼近理想解排序法),构建了CW-TOPSIS煤矿内因火灾评价模型,评价火灾危险等级。运用该模型对新安煤矿火灾等级评价,危险等级为"较安全",指标对煤矿内因火灾的影响由大到小为:煤层埋深与地温>采煤方法及工艺>开拓方式及布置>干煤吸氧量>煤层含水量>灰分含量,评价结果与工程实际吻合。该模型的评价结果与传统内因火灾评价方法 -BP神经网络法、模糊综合评价法、层次分析法相比较,CW-TOPSIS评价模型精度较高、更可靠,具有较高的应用价值。

Coal mine internal cause of fire evaluation index has the uncertainty of gray and unknown,so it is difficult to determine the reasonable index weight,which leads to the inaccuracy of the evaluation.To solve this problem,we built a coal mine internal-caused fire evaluation system based on the standard layer four factors,including coal seam geological storage,exploitation and mining technology,coal seam spontaneous combustion tendency and ventilation condition and goaf management,and the index layer 24 factors,the thickness of coal seam,the dip angle of the coal seam and the lithology of the coal seam;put forward a method of combination weighting,which combines AHP(analytic hierarchy process)the subjective with objective weighting method of entropy,introduced Lagrange function,and constructed the optimal decision model.Through the Euclidean distance function,maintained the consistency of subjective and objective weights and the preference coefficient,and obtained index weights and the relationship between primary and secondary for evaluation index.By use of TOPSIS(TOPSIS),constructed the CW-TOPSIS internal fire evaluation model for evaluation of fire danger rating.Evaluated a coal mine fire risk grade by using this model,and its danger degree is“safer”.Indices on the internal combustion coal mine fire descending order as follows:coal seam depth and ground temperature>Coal mining methods and processes>development and layout>dry coal oxygen absorption>coal seam water content>ash content.The evaluation result agrees with the project.Compared with BP neural network method,fuzzy comprehensive evaluation method and analytic hierarchy process,the CW-TOPSIS evaluation model has higher accuracy with higher application value.