贝叶斯判别和Fisher判别在矿井突水水源判别中的应用比较  

Application comparison of Bayes and Fisher discrimination in mine inrush water source

作　　者：于小鸽 裴富华 刘燚菲 衡培国 刘延 吕伟魁 YU Xiaoge;PEI Fuhua;LIU Yifei;HENG Peiguo;LIU Yan;LÜ Weikui(College of Resources, Shandong University of Science and Technology, Taian, Shandong 271019, China;College of Earth Science and Engineering, Shandong University of Science and Technology, Qingdao, Shandong 266590, China;Guhanshan Mine, Jiaozuo Coal Industry (Group) Co. , Ltd. , Jiaozuo, Henan 454000, China;Shandong New Julong Energy Co. , Ltd. , Heze, Shandong 273700, China)

机构地区：[1]山东科技大学资源学院,山东泰安271019 [2]山东科技大学地球科学与工程学院,山东青岛266590 [3]焦作煤业(集团)有限责任公司古汉山矿,河南焦作454000 [4]山东新巨龙能源有限责任公司,山东菏泽273700

出　　处：《中国科技论文》2022年第1期9-14,共6页China Sciencepaper

基　　金：国家自然科学基金资助项目(42002282)。

摘　　要：矿井突水水源的有效识别对于预防矿井突水事故起到至关重要的作用,为了解决传统水源判别方法存在准确率较低、所需数据量大的问题,依据山东盛泉煤矿突水水源的水质资料,选取Na^(+)+K^(+)、Ca^(2+)、Mg^(2+)、Cl^(-)、SO^(2-)_(4)和HCO^(-)_(3)这6种水化学成分指标作为矿井突水水源判别的依据,分别构建Fisher线性判别模型和朴素贝叶斯分类模型,并将2种模型的结果进行比较。结果表明,2种方法对于矿井突水水源的识别均有一定的实用性,但利用MATLAB实现的贝叶斯判别函数更为简单,且运行速度较快,误判率较低,因而可作为多分类矿井突水水源判别的有效手段。Effective identification of mine water inrush source plays a vital role in preventing mine water inrush accident.The traditional identification method of mine water source has the characteristics of low accuracy and large amount of data.In order to solve this problem,six kinds of water chemical composition indexes,Na^(+)+K^(+),Ca^(2+),Mg^(2+),Cl^(-),SO^(2-)_(4) and HCO^(-)_(3),were selected as the discriminant basis of mine water inrush source based on the water quality data of Shengquan coal mine,and the Fisher linear discriminant model and naive Bayesian classification model were conctructed respectively.By comparing the two models,it is found that the two methods have certain practicability for the identification of mine water inrush source,but the Bayesian discriminant function realized by MATLAB software is simpler,with faster running speed and lower misjudgment rate,so it can be used as an effective means for the identification of multi-classification of mine water inrush source.

关 键 词：突水水源识别 Fisher线性判别模型 朴素贝叶斯分类模型 SPSS软件 MATLAB软件 

分 类 号：TD745[矿业工程—矿井通风与安全]