基于DILSSVM的岩石强度预测研究
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
为解决经典强度准则预测岩石破坏强度不准确的问题,本文使用交叉最小二乘支持向量机(DILSSVM)在单轴及三轴加载情况下对岩石强度进行预测。将现场和实验室试验得到的各种岩石相关数据划分成训练和验证子集。将抗压强度σc和最小主应力σ3作为输入值,最大主应力值σ1f作为输出值训练DILSSVM。使用训练后的DILSSVM预测岩石破坏时的σ1f,同时使用训练子集反演Hoek-Brown经验公式的常数m。使用测试数据对训练后DILSSVM预测的目标岩石强度的准确性进行验证,同时将测试数据带入2类Hoek-Brown(m值的取值不同)经验公式预测目标岩石强度σ1f。比较DILSSVM和HoekBrown预测结果,表明:DILSSVM预测结果的均方差减小了45%~55%,决定系数增加了0.055~0.085,更接近于1。说明DILSSVM对岩石强度的预测范围较宽,且适合岩石种类多变的复杂非线性情况。
The classical failure criteria for prediction of rock strength can not be accurate. This study uses the diagonal intersection least squares SVM( DILSSVM) as new method for prediction of the rock strength in a wide loading such as uniaxial and triaxial loading. For each rock type,data obtained from field experiments and laboratory experiments are divided into training and test sets. DILSSVM is employed to train with the compressive stress( σc) and minor principal stress( σ3) and to predict the value of major principal stress( σ1f) at failure. The training sets are used in regression analysis for m in Hoek-Brown( H-B) equation. Then,the test sets are used to examine the accuracy of target rock strength with DILSSVM after training and these of the two H-B model( with different m). Comparison of the results of the DILSSVM with the two H-B models shows that the DILSSVM alwayshas less root mean squared error( decreased 45% ~ 55%) and higher coefficient of determination( enhanced0. 055 ~ 0. 085,near to 1). The DILSSVM shows better flexibility in σ1fat failure in each rock type and a wide loading range.
引文
[1]吴黎辉.岩体经验强度准则研究[硕士学位论文][D].西安:长安大学,2004.Wu Lihui.Research on rock mass experience strength criterion[Thesis of Master][D].Xi'an:Chang'an University,2004.
    [2]崔铁军,马云东,肖晓春.基于神经网络的岩石强度准则研究[J].华中师范大学学报(自然科学版),2014,48(1):131~135.Cui Tiejun,Ma Yundong,Xiao Xiaochun.Research on the rock strength based on the artificial neural networks[J].Journal of Huazhong Normal University(Natural Science),2014,48(1):131~135.
    [3]汤献良,赵建军,黄润秋.小湾水电站坝基岩体质量动态评价研究[J].工程地质学报,2013,21(3):370~376.Tang Xianliang,Zhao Jianjun,Huang Runqiu.Dynamic evaluation of rock mass quality for dam foundation at Xiaowan hydropower station[J].Journal of Engineering Geology,2013,21(3):370~376.
    [4]许飞,胡修文,黄香亮,等.边坡岩体力学参数对Hoek-Brown准则参数敏感性的综合性分析[J].工程地质学报,2013,21(4):613~618.Xu Fei,Hu Xiuwen,Huang Xiangliang,et al.Orthogonal analysis of sensitivity of mechanical parameters of slope rock mass to parameters in Hoek-Brown criterion[J].Journal of Engineering Geology,2013,21(4):613~618.
    [5]Vapnik V N.The nature of statistical learning theory[M].New York,USA:Springer-Verlag,1995.
    [6]Vapnik V,Levin E,Cun Y L.Measuring the VC-dimension of a learning machine[J].Neural Computation,1994,6(5):851~876.
    [7]Singh V K,Singh D,Singh T N.Prediction of strength properties of some schistose rocks from petrographic properties using artificial neural networks[J].International Journal of Rock Mechanics and Mining Sciences,2001,38(2):269~284.
    [8]Demuth H,Beal M,Hagan M.Neural network toolbox 5 user's guide of the MATLAB software[S].Natick,Mass,2007.
    [9]Hecht-Nielsen R.Kolmogorov's mapping neural network existence theorem[A]∥Proceedings of the First IEEE International Conference on Neural Networks[C].San Diego:1987,11~15.
    [10]Aldrich C,van Deventer J S J,Reuter M A.The application of neural nets in the metallurgical industry[J].Minerals Engineering,1994,7(5—6):793~809.
    [11]崔铁军,马云东,白润才.基于神经网络优化遗传算法的爆破参数优化[J].地震工程与工程振动,2014,34(1):257~262.Cui Tiejun,Ma Yundong,Bai Runcai.Optimization of blast parameters using genetic algorithm optimized by artificial neural network[J].Earthquake Engineering and Engineering Dynamics,2014,34(1):257~262.
    [12]Hoek E,Brown ET.Underground excavations in rock[M].London:Institution of Mining and Metallurgy,1980.
    [13]张后全,徐建峰,贺永年,等.脆性岩石真三轴能量强度准则研究[J].中国矿业大学学报,2012,41(4):564~570.Zhang Houquan,Xu Jianfeng,He Yongnian,et al.Study of triaxial energy strength criterion for brittle rock materials[J].Journal of China University of Mining&Technology,2012,41(4):564~570.
    [14]王洪德,崔铁军,马云东.基于差异进化算法的爆破震速参数识别与优化[J].中国安全科学学报,2012,22(5):17~23.Wang Hongde,Cui Tiejun,Ma Yundong.Blasting vibration velocity parameters recognition and optimization based on differential evolution algorithm[J].China Safety Science Journal,2012,22(5):17~23.
    [15]崔铁军,马云东.基于差异进化支持向量机的坑外土体沉降预测[J].中国安全科学学报,2013,23(1):83~89.Cui Tiejun,Ma Yundong.Prediction of soil settlement outside pit based on DE and SVM[J].China Safety Science Journal,2013,23(1):83~89.
    [16]Everitt B S.The cambridge dictionary of statistics[M].New York:Cambridge University Press,2006.

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