ISSN 1000-3665 CN 11-2202/P
  • 中文核心期刊
  • Scopus 收录期刊
  • 中国科技核心期刊
  • DOAJ 收录期刊
  • CSCD(核心库)来源期刊
  • 《WJCI 报告》收录期刊
欢迎扫码关注“i环境微平台”

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于无监督方法确定岩土参数取值

阮永芬 李鹏辉 朱强 王勇 闫明

阮永芬,李鹏辉,朱强,等. 基于无监督方法确定岩土参数取值[J]. 水文地质工程地质,2023,50(0): 1-11 doi:  10.16030/j.cnki.issn.1000-3665.202207046
引用本文: 阮永芬,李鹏辉,朱强,等. 基于无监督方法确定岩土参数取值[J]. 水文地质工程地质,2023,50(0): 1-11 doi:  10.16030/j.cnki.issn.1000-3665.202207046
RUAN Yongfen, LI Penghui, ZHU Qiang, et al. Determination of geotechnical parameters based on the unsupervised learning method[J]. Hydrogeology & Engineering Geology, 2023, 50(0): 1-11 doi:  10.16030/j.cnki.issn.1000-3665.202207046
Citation: RUAN Yongfen, LI Penghui, ZHU Qiang, et al. Determination of geotechnical parameters based on the unsupervised learning method[J]. Hydrogeology & Engineering Geology, 2023, 50(0): 1-11 doi:  10.16030/j.cnki.issn.1000-3665.202207046

基于无监督方法确定岩土参数取值

doi: 10.16030/j.cnki.issn.1000-3665.202207046
基金项目: 中铁二十局集团第五工程有限公司科研计划项目(CR2005-5-JS-2021-009)
详细信息
    作者简介:

    阮永芬(1964-),女,博士,教授,主要从事岩土工程方面的研究。E-mail:rryy64@163.com

    通讯作者:

    李鹏辉(1999-),男,硕士研究生,主要从事岩土工程方面的研究。E-mail:1017343481@qq.com

  • 中图分类号: TU443

Determination of geotechnical parameters based on the unsupervised learning method

  • 摘要: 随着城市工程建设的发展,建筑工程事故问题愈发突出,采用传统方法求取的岩土参数区间无法满足实际工程需要。基于无监督学习思想,选取工程性质较差的泥炭质土,结合工程经验选用8个物理指标作为输入集,利用主成分分析(PCA)算法实现多样本多参数去耦合的降维处理,得出各物理指标相关性及敏感度,结合其相关性及敏感度赋予不同埋深泥炭质土物理指标的综合评价值。利用k-means聚类分析泥炭质土物理指标、综合评价值及工程特性之间关系,为岩土参数选取提供理论基础。采用监督学习方法——BP神经网络算法分析无监督结果,验证(PCA—k-means)算法模型的合理性。将通过聚类分析得到的正态样本利用多种截尾法优化,得到可靠取值区间,并将取值结果与实际工程取值比较,验证该模型工程参数取值合理性。该算法模型具有较好的工程应用价值,所得研究结果可为工程勘察、设计、施工参数取值提供参考,也能为岩土参数取值分析提供新的分析方法。
  • 图  1  会展中心地层构造图

    Figure  1.  Stratigraphic diagram of the convention center

    图  2  ρwwLea1-2关系曲线

    Figure  2.  Relationship of ρ and w, wL, e and a1-2

    图  3  wuwwLea1-2关系曲线

    Figure  3.  Relationship of wu and w, wL, e and a1-2

    图  4  特征值变化规律

    Figure  4.  Variation of elgenvalues

    图  5  累计贡献率

    Figure  5.  Cumulative contribution

    图  6  判别指标分类结果

    Figure  6.  Discriminative index classification results

    图  7  综合评价分类结果

    Figure  7.  Results of comprehensive evaluation

    图  8  ewwL分类结果

    Figure  8.  Classification results of ew and wL

    图  9  分类结果对比

    Figure  9.  Comparison of classification results

    图  10  综合评价值对比

    Figure  10.  Comprehensive evaluation and comparison

    图  11  3个物理指标原样本分布图

    Figure  11.  Original sample distribution map of three physical indicators

    图  12  分类后3个物理指标样本分布图

    Figure  12.  Sample distribution map of three physical indicators after classification

    表  1  各层泥炭质土的物理指标统计表

    Table  1.   Statistical table of physical indexes of peaty soil in each layer

    地层编号ρ
    /(g·cm−3
    Gsw
    /%
    wL
    /%
    wP
    /%
    ea1-2
    /MPa−1
    wu
    /%
    $\text{③}_1^1$1.612.30110.1135.4100.02.491.6224.33
    $\text{③}_2 $1.422.42101.2119.287.22.401.3218.36
    $\text{④}_1 $1.251.95145.7182.1146.42.931.4940.10
    $\text{⑤}_{12}$1.271.97130.6174.3140.62.521.1238.77
    下载: 导出CSV

    表  2  不同物理指标间的相关性系数

    Table  2.   Correlation between different physical indicators

    R,2ρGSwwLea1-2wPwu
    ρ1.00
    GS0.641.00
    w0.870.551.00
    wL0.890.590.941.00
    e0.850.850.940.871.00
    a1-20.620.420.810.680.851.00
    wP0.850.580.820.930.580.491.00
    wu0.830.660.730.780.700.450.801.00
    下载: 导出CSV

    表  3  各物理指标敏感度统计表

    Table  3.   Sensitivity statistics of each physical index

    物理指标编码X1(ρ)X2GSX3wX4wLX5eX6a1-2X7wPX8wu
    敏感度0.23490.20690.25400.24280.26080.21510.24100.2281
    下载: 导出CSV

    表  4  各组泥炭质土综合评价值排序

    Table  4.   Ranking of comprehensive evaluation of the peaty soil

    组类ρ/(g·cm−3w/%GSewL/%wP/%a1-2/MPa-1wu/%综合评价
    11.56557.5902.5171.53668.69042.6900.7228.100–1.295
    21.53367.9802.3971.56571.73044.9100.81212.670–1.162
    31.49968.2802.4281.69976.18048.2701.10311.690–1.096
    41.48372.1202.4361.82484.80049.5600.97712.270–1.035
    51.45273.6902.2381.88292.40058.8900.85013.678–0.928
    61.45478.0702.4261.97190.32053.5501.12112.100–0.946
    71.45582.1802.4261.95795.36362.1810.83313.318–0.900
    341.115211.2001.7964.406228.270149.4003.80243.7501.293
    351.134220.2001.1644.761228.710144.1004.71844.3461.496
    361.117232.7001.3114.818238.770149.3105.03140.8171.566
    371.121247.0000.9965.327249.400156.3006.00641.7101.861
    381.079287.1820.9905.839304.936207.8005.97856.4552.574
    下载: 导出CSV

    表  5  不同截尾法的原物理指标区间范围

    Table  5.   Original physical index range of different censoring methods

    物理指标均值标准差偏度系数变异系数3σ区间范围c33区间范围c3区间范围
    e 3.031.200.940.39[0,6.63][0.56,7.76][0,7.76]
    w/%133.0262.090.930.46[0,319.29][4.48,377.03][0,377.03]
    wL/%148.6361.260.870.41[0,332.41][18.14,385.71][0,385.71]
    下载: 导出CSV

    表  6  物理指标样本检验结果

    Table  6.   Normal test result of the physical index sample

    指标分布类型D值P值检验结果
    e图11正态分布0.1551.81×10–4排除
    e图12正态分布0.0901.00接受
    w图11正态分布0.1892.36×10–12排除
    w图12正态分布0.0640.710接受
    wL图11正态分布0.1461.62×10–7排除
    wL图12正态分布0.0720.190接受
    下载: 导出CSV

    表  7  不同截尾法的新样本区间范围

    Table  7.   Range of new sample intervals for different censoring methods

    物理指标均值标准差偏度系数变异系数3σ区间范围c33区间范围c3区间范围
    e2.190.400.390.18[0.99,3.39][1.45,3.55][0.99,3.55]
    w/%88.6817.400.390.18[36.48,140.88][43.26,147.66][36.48,140.88]
    wL/%105.5522.110.280.20[39.22,171.88][45.41,178.07][39.22,178.08]
    ρ/(g·cm−31.400.080.080.06[1.13,1.66][1.15,1.67][1.13,1.67]
    Gs2.250.45–3.080.20[0.89,3.61][1.15,3.55][1.15,3.61]
    a1-2/MPa−11.240.561.560.45[0,2.93][0.43,3.81][0,2.93]
    wP/%68.7618.060.380.26[14.57,122.97][21.48,129.88][14.57,129.88]
    wu/%17.637.642.020.43[0,40.57][10.17,56.06][0,56.06]
    下载: 导出CSV

    表  8  各层泥炭质土设计参数检验结果

    Table  8.   Test results of design parameters of the peaty soil in each layer

    地层
    编号
    ρ
    /(g·cm−3
    w
    /%
    eILa1-2
    /MPa−1
    综合
    评价
    所属类
    $ \text{③}_1^1$1.37103.02.410.151.63−0.23第一类
    $\text{③}_2 $1.3991.12.330.121.30−0.26第一类
    $\text{④}_1 $1.20142.32.92−0.141.47−0.08第一类
    $\text{⑤}_{12} $1.26127.72.53−0.281.12−0.16第一类
    下载: 导出CSV
  • [1] 阮永芬,刘岳东,王东,等. 昆明泥炭与泥炭质土对建筑地基的影响[J]. 昆明理工大学学报(理工版),2003,28(3):121 − 124. [RUAN Yongfen,LIU Yuedong,WANG Dong,et al. Effect of Kunming’s peat & peaty soil on the building foundation[J]. Journal of Kunming University of Science and Technology,2003,28(3):121 − 124. (in Chinese with English abstract)
    [2] 柴春阳,张广泽,胡婷. 丽香铁路冰水沉积型泥炭土特性及工程对策分析[J]. 地下空间与工程学报,2017,13(增刊 2):757 − 761. [CHAI Chunyang,ZHANG Guangze,HU Ting. Characteristics of ice-water deposition peat in Li-Xiang railway and analysis on engineering countermeasures[J]. Chinese Journal of Underground Space and Engineering,2017,13(Sup 2):757 − 761.
    [3] 李强,李刚. 沪苏浙高速公路泥炭质土的工程性质研究[J]. 公路工程,2010,35(6):99 − 103. [LI Qiang,LI Gang. Research on the engineering property of peat quality soil in hu-su-Zhe highway[J]. Highway Engineering,2010,35(6):99 − 103. doi:  10.3969/j.issn.1674-0610.2010.06.026
    [4] 谢俊文. 泥炭的工程地质特性[J]. 工程勘察,1981,9(5):31 − 34. [XIE Junwen. Engineering geological characteristics of peat[J]. Geotechnical Investigation & Surveying,1981,9(5):31 − 34. (in Chinese with English abstract)
    [5] KOLAY P K, AMINUR M R, TAIB S N L, et al. Correlation between different physical and engineering properties of tropical peat soils from Sarawak[C]//Soil Behavior and Geo-Micromechanics. Reston, VA: American Society of Civil Engineers, 2010: 56-61.
    [6] PRICE J S,CAGAMPAN J,KELLNER E. Assessment of peat compressibility:is there an easy way?[J]. Hydrological Processes,2005,19(17):3469 − 3475. doi:  10.1002/hyp.6068
    [7] SANTAGATA M,BOBET A,JOHNSTON C T,et al. One-dimensional compression behavior of a soil with high organic matter content[J]. Journal of Geotechnical and Geoenvironmental Engineering,2008,134(1):1 − 13. doi:  10.1061/(ASCE)1090-0241(2008)134:1(1)
    [8] 裴利华,杨醒宇,桂跃,等. 有机质含量及组分对泥炭土物理力学性质影响[J]. 水文地质工程地质,2022,49(2):77 − 85. [PEI Lihua,YANG Xingyu,GUI Yue,et al. Influence of organic matter content and ingredient on the physical and mechanical properties of peat soils[J]. Hydrogeology & Engineering Geology,2022,49(2):77 − 85. (in Chinese with English abstract)
    [9] 严春风,刘东燕,朱可善. 岩石抗剪强度指标互相关性的敏感度分析[J]. 水文地质工程地质,1997,24(6):45 − 47. [YAN Chunfeng,LIU Dongyan,ZHU Keshan. Susceptibility analysis tothe correlativity of shear strength indices of ricks[J]. Hydrogeology & Engineering Geology,1997,24(6):45 − 47. (in Chinese with English abstract)
    [10] 程豪,田林. 昆明市某场地不同埋深下泥炭质土物理力学指标统计规律研究[J]. 贵州大学学报(自然科学版),2022,39(4):92 − 98. [CHENG Hao,TIAN Lin. Statistical law of physical and mechanical indexes of peat soil under different burial depths in a site in Kunming[J]. Journal of Guizhou University (Natural Science),2022,39(4):92 − 98. (in Chinese with English abstract)
    [11] 丁祖德,付江,李夕松,等. 昆明泥炭质土的物理力学指标特征及相关性分析[J]. 公路工程,2018,43(4):86 − 91. [DING Zude,FU Jiang,LI Xisong,et al. Properties of physical and mechanical indexes and correlation analysis of peaty soil in Kunming area[J]. Highway Engineering,2018,43(4):86 − 91. (in Chinese with English abstract)
    [12] 陈玉萍,袁志强,周博,等. 遗传算法优化BP网络在滑坡灾害预测中的应用研究[J]. 水文地质工程地质,2012,39(1):114 − 119. [CHEN Yuping,YUAN Zhiqiang,ZHOU Bo,et al. ] Application of back propagation neural networks with optimization of genetic algorithms to landslide hazard prediction[J]. Hydrogeology & Engineering Geology,2012,39(1):114 − 119. (in Chinese with English abstract)
    [13] 马壮壮,束龙仓,季叶飞,等. 基于遗传算法的BP神经网络计算岩溶水安全开采量[J]. 水文地质工程地质,2016,43(1):22 − 27. [MA Zhuangzhuang,SHU Longcang,JI Yefei,et al. Calculation of Karst water safe yield by using BP neural network based on genetic algorithm[J]. Hydrogeology & Engineering Geology,2016,43(1):22 − 27. (in Chinese with English abstract)
    [14] 阮永芬, 张虔, 乔文件, 等. 盾构施工地表沉降的无监督学习预估方法[J/OL]. 安全与环境学报. (2022-05-09)[2022-09-05]. DOI:  10.13637/j.issn.1009-6094.2022.0351.

    RUAN Yongfen, ZHANG Qian, QIAO Wenjian, et al. Unsupervised learning prediction method of ground settlement in shield construction[J/OL]. Journal of Safety and Environment. (2022-05-09)[2022-09-05]. (in Chinese with English abstract)
    [15] 宫凤强,鲁金涛. 基于主成分分析与距离判别分析法的突水水源识别方法[J]. 采矿与安全工程学报,2014,31(2):236 − 242. [GONG Fengqiang,LU Jintao. Recognition method of mine water inrush sources based on the principal element analysis and distance discrimination analysis[J]. Journal of Mining & Safety Engineering,2014,31(2):236 − 242. (in Chinese with English abstract)
    [16] 任玉晓. 地震波速无监督深度学习反演方法及其在隧道超前探测中的应用[D]. 济南: 山东大学, 2021

    REN Yuxiao. Unsupervised deep-learning inversion method for seismic velocity and its application in tunnel forward-prospecting[D]. Jinan: Shandong University, 2021. (in Chinese with English abstract)
    [17] 阮永芬,魏德永,高骏,等. K-means聚类分析高原湖相沉积软土参数[J]. 昆明理工大学学报(自然科学版),2020,45(1):85 − 91. [RUAN Yongfen,WEI Deyong,GAO Jun,et al. K-means cluster analysis of soft soil parameters of plateau sedimentary[J]. Journal of Kunming University of Science and Technology (Natural Science),2020,45(1):85 − 91. (in Chinese with English abstract)
    [18] 孔令奇,李翠娟. 岩土参数概率分布的系统推断研究[J]. 工业建筑,2022,52(1):129 − 136. [KONG Lingqi,LI Cuijuan. Study on system inference of probability distribution of geotechnical parameters[J]. Industrial Construction,2022,52(1):129 − 136. (in Chinese with English abstract)
    [19] 殷瑞刚,魏帅,李晗,等. 深度学习中的无监督学习方法综述[J]. 计算机系统应用,2016,25(8):1 − 7. [YIN Ruigang, WEI Shuai, LI Han, et al. Introduction of unsupervised learning methods in deep learning[J]. Computer Systems & Applications,2016,25(8):1 − 7. (in Chinese with English abstract)
    [20] 孙吉贵,刘杰,赵连宇. 聚类算法研究[J]. 软件学报,2008,19(1):48 − 61. [SUN Jigui,LIU Jie,ZHAO Lianyu. Clustering algorithms research[J]. Journal of Software,2008,19(1):48 − 61. (in Chinese with English abstract) doi:  10.3724/SP.J.1001.2008.00048
    [21] 张麟, 潘红岩. 聚类分析算法应用研究[J]. 数字技术与应用, 2016(10): 143+145

    ZHANG Lin, PAN Hongyan. Research on application of clustering analysis algorithm[J]. Digital Technology and Application, 2016(10): 143+145. (in Chinese)
    [22] 刘荣. 人工神经网络基本原理概述[J]. 计算机产品与流通, 2020(6): 35+81

    LIU Rong. Overview of basic principles of artificial neural network[J]. Computer Products & Circulation, 2020(6): 35+81. (in Chinese)
    [23] 袁冰清,程功,郑柳刚. BP神经网络基本原理[J]. 数字通信世界,2018(8):28 − 29. [YUAN Bingqing,CHENG Gong,ZHENG Liugang. Basic principle of BP neural networks[J]. Digital Communication World,2018(8):28 − 29. (in Chinese with English abstract)
    [24] 宫凤强,侯尚骞,李夕兵. 岩土参数截尾分布的正态信息扩散推断方法[J]. 武汉大学学报(工学版),2016,49(5):661 − 667. [GONG Fengqiang,HOU Shangqian,LI Xibing. Truncated distribution deduction method for geotechnical parameters based on normal information diffusion method[J]. Engineering Journal of Wuhan University,2016,49(5):661 − 667. (in Chinese with English abstract)
  • 加载中
图(12) / 表(8)
计量
  • 文章访问数:  3755
  • HTML全文浏览量:  129
  • PDF下载量:  45
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-07-28
  • 修回日期:  2022-10-20
  • 网络出版日期:  2023-04-03

目录

    /

    返回文章
    返回