松动圈厚度预测的支持向量机模型
|
摘要
为将支持向量机理论应用于浅埋巷道围岩松动圈厚度的预测中,结合巷道矿压显现特点和松动圈支护理论,选取巷道埋深、巷道跨度、围岩强度、围岩节理裂隙发育程度和巷道掘进断面积等5个主要影响因素作为松动圈厚度的判别指标。根据支持向量机回归预测算法,以20组实测数据作为训练样本,基于Libsvm软件建立了松动圈厚度预测的支持向量机模型。为检验该模型的可靠性,利用该模型对10组测试样本的松动圈厚度进行预测并与实测值做了对比分析。结果表明,用该模型预测松动圈厚度的平均相对误差为2.66%,比神经网络方法和灰色理论方法预测的结果更精确。这充分说明用支持向量机模型预测松动圈厚度具有良好的实用性和较高的可靠度,可为松动圈厚度的确定提供一条新的途径。
To apply support vector machine(SVM) theory to predict the thickness of excavation damaged zone(EDZ) of shallow roadway,the depth,span,intensity of rock,rock joint development degree,and roadway basal area are regarded as the discriminant factors of the thickness of EDZ,with a consideration of characteristics of underground pressure and the support theory of EDZ.The SVM model for predicting the thickness of EDZ is established by using 20 groups of measured data as training samples,based on Libsvm software and the regression prediction algorithm of SVM.To test the model's reliability,10 groups of measured data are used as testing samples and the predictive value by the model is compared with the measured value.The results show that the model's average relative error is 2.66%,which is more accurate than that by the methods of artificial neural network and gray theory,showing that the model is practical and reliable,and could provide a new way for ascertaining the thickness of EDZ.
To apply support vector machine(SVM) theory to predict the thickness of excavation damaged zone(EDZ) of shallow roadway,the depth,span,intensity of rock,rock joint development degree,and roadway basal area are regarded as the discriminant factors of the thickness of EDZ,with a consideration of characteristics of underground pressure and the support theory of EDZ.The SVM model for predicting the thickness of EDZ is established by using 20 groups of measured data as training samples,based on Libsvm software and the regression prediction algorithm of SVM.To test the model's reliability,10 groups of measured data are used as testing samples and the predictive value by the model is compared with the measured value.The results show that the model's average relative error is 2.66%,which is more accurate than that by the methods of artificial neural network and gray theory,showing that the model is practical and reliable,and could provide a new way for ascertaining the thickness of EDZ.
引文
[1]董方庭.巷道围岩松动圈支护理论与应用技术[M].北京:煤炭工业出版社,2001.
[2]孟庆彬,乔卫国,林登阁,等.地质雷达测试巷道围岩松动圈的原理及应用[J].矿业安全与环保,2011,38(2),37-42.
[3]赵海斌,付建军,周江平.高地应力区大型地下洞室边墙松动圈动态分析研究[J].长江科学院院报,2011,28(9),35-39.
[4]孙有为.地下洞室的几何性质对松动圈的影响[D].哈尔滨:中国地震局工程力学研究所,2006.
[5]万串串,李夕兵,马春德.基于围岩松动圈现场测试的深部软岩巷道支护技术优化[J].矿冶工程,2012,32(1),12-16.
[6]周志华,叶洲元,段瑜.软岩巷道围岩松动圈的数值模拟[J].中国矿业,2009,18(10),97-99.
[7]薛新华.遗传神经网络法在巷道围岩松动圈预测中的应用[J].岩土工程技术,2006,20(5),237-239,266.
[8]檀远远.复杂构造带回采巷道松动圈确定与支护对策研究[D].淮南:安徽理工大学能源与安全学院,2009.
[9]李政林,吴瑞祥,李龙剑,等.基于损伤理论的隧道围岩松动圈确定方法[J].地下空间与工程学报,2011,7(6),1060-1064,1090.
[10]钱七虎,李树忱.深部岩体工程围岩分区破裂化现象研究综述[J].岩石力学与工程学报,2008,27(6),1278-1284.
[11]谢理,张认成,蔡钟山.支持向量机在火灾早期探测中的应用[J].广西大学学报:自然科学版,2011,36(2):246-250.
[12]KARIM O E,MAHMULOD O E.Predicting defect-prone software modules using support vector machines[J].Journal of Systems and Software,2008,81(5),649-660.
[13]谭锴.基于支持向量机的云南地区地震预测[D].杭州:浙江师范大学数理与信息工程学院,2010.
[14]CAI M,KAISER P K.Assessment of excavation damaged zone using a micromechanics model[J].Tunnelling and Under-ground Space Technology,2005,20(4),301-310.
[15]高玮,郑颖人.巷道围岩松动圈预测的进化神经网络法[J].岩石力学与工程学报,2002,21(5),658-661.
[16]朱川曲.回采巷道围岩稳定性分类及松动圈尺寸预测[J].黄金科学技术,1999,7(45),63-66.
[2]孟庆彬,乔卫国,林登阁,等.地质雷达测试巷道围岩松动圈的原理及应用[J].矿业安全与环保,2011,38(2),37-42.
[3]赵海斌,付建军,周江平.高地应力区大型地下洞室边墙松动圈动态分析研究[J].长江科学院院报,2011,28(9),35-39.
[4]孙有为.地下洞室的几何性质对松动圈的影响[D].哈尔滨:中国地震局工程力学研究所,2006.
[5]万串串,李夕兵,马春德.基于围岩松动圈现场测试的深部软岩巷道支护技术优化[J].矿冶工程,2012,32(1),12-16.
[6]周志华,叶洲元,段瑜.软岩巷道围岩松动圈的数值模拟[J].中国矿业,2009,18(10),97-99.
[7]薛新华.遗传神经网络法在巷道围岩松动圈预测中的应用[J].岩土工程技术,2006,20(5),237-239,266.
[8]檀远远.复杂构造带回采巷道松动圈确定与支护对策研究[D].淮南:安徽理工大学能源与安全学院,2009.
[9]李政林,吴瑞祥,李龙剑,等.基于损伤理论的隧道围岩松动圈确定方法[J].地下空间与工程学报,2011,7(6),1060-1064,1090.
[10]钱七虎,李树忱.深部岩体工程围岩分区破裂化现象研究综述[J].岩石力学与工程学报,2008,27(6),1278-1284.
[11]谢理,张认成,蔡钟山.支持向量机在火灾早期探测中的应用[J].广西大学学报:自然科学版,2011,36(2):246-250.
[12]KARIM O E,MAHMULOD O E.Predicting defect-prone software modules using support vector machines[J].Journal of Systems and Software,2008,81(5),649-660.
[13]谭锴.基于支持向量机的云南地区地震预测[D].杭州:浙江师范大学数理与信息工程学院,2010.
[14]CAI M,KAISER P K.Assessment of excavation damaged zone using a micromechanics model[J].Tunnelling and Under-ground Space Technology,2005,20(4),301-310.
[15]高玮,郑颖人.巷道围岩松动圈预测的进化神经网络法[J].岩石力学与工程学报,2002,21(5),658-661.
[16]朱川曲.回采巷道围岩稳定性分类及松动圈尺寸预测[J].黄金科学技术,1999,7(45),63-66.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心