回归分析方法对比及在瓦斯赋存规律中的应用  

Regression Analysis Method of Contrast Gas Occurrence Regularity Application

作　　者：李昊龙[1] 

机构地区：[1]河南理工大学安全科学与工程学院,河南焦作454000

出　　处：《煤》2016年第3期17-19,53,共4页Coal

摘　　要：为进一步提高矿井瓦斯预测的精确度,在现有矿井瓦斯预测多元线性回归分析的基础上,对多元非线性回归分析方法做进一步探索。通过利用Excel中的数值拟合模型结合裴沟煤矿樊寨井田实测的瓦斯参数,对影响该井田瓦斯赋存的地质因素进行分析,得到影响该井田瓦斯赋存的主要地质因素为埋深、滑动面距底板距离和含砂率;并通过对单因素线性、单因素非线性、多因素线性和多因素非线性拟合曲线的回归分析结果之间的对比,得到多因素非线性拟合曲线数值拟合回归效果最好。对瓦斯赋存预测的研究有一定的指导作用。In order to further improve the accuracy of prediction of mine gas,further exploration about multivariate nonlinear regression analysis method was made,which is based on existing gas forecast multiple linear regression analysis in the mine. Through using the numerical fitting model in Excel combined with the gas parameters of Peigou Coal Mine in Fanzhai minefield,we analysed the geological factors affecting the deposit of mine gas. Then we got the conclusion that the major geological factors of gas deposit are depth of the mine,sliding plane distance from the base plate and sand ratio. Through the comparative analysis among results of curve fitting regression for the single factor linear,single factor nonlinear,multivariate factor linear and multivariate factor nonlinear,we can know that curve fitting regression for multivariate nonlinear has the best numerical fitting effect. It will guide the research of forecast about gas deposit to a certain degree.

关 键 词：瓦斯赋存 回归分析 线性 非线性 

分 类 号：TD712.5[矿业工程—矿井通风与安全]