• 中国出版政府奖提名奖

    中国百强科技报刊

    湖北出版政府奖

    中国高校百佳科技期刊

    中国最美期刊

    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

    基于集成学习与贝叶斯优化的岩石抗压强度预测

    吴禄源 李建会 马丹 王自法 张建伟 袁超 冯义 李辉

    吴禄源, 李建会, 马丹, 王自法, 张建伟, 袁超, 冯义, 李辉, 2023. 基于集成学习与贝叶斯优化的岩石抗压强度预测. 地球科学, 48(5): 1686-1695. doi: 10.3799/dqkx.2023.029
    引用本文: 吴禄源, 李建会, 马丹, 王自法, 张建伟, 袁超, 冯义, 李辉, 2023. 基于集成学习与贝叶斯优化的岩石抗压强度预测. 地球科学, 48(5): 1686-1695. doi: 10.3799/dqkx.2023.029
    Wu Luyuan, Li Jianhui, Ma Dan, Wang Zifa, Zhang Jianwei, Yuan Chao, Feng Yi, Li Hui, 2023. Prediction for Rock Compressive Strength Based on Ensemble Learning and Bayesian Optimization. Earth Science, 48(5): 1686-1695. doi: 10.3799/dqkx.2023.029
    Citation: Wu Luyuan, Li Jianhui, Ma Dan, Wang Zifa, Zhang Jianwei, Yuan Chao, Feng Yi, Li Hui, 2023. Prediction for Rock Compressive Strength Based on Ensemble Learning and Bayesian Optimization. Earth Science, 48(5): 1686-1695. doi: 10.3799/dqkx.2023.029

    基于集成学习与贝叶斯优化的岩石抗压强度预测

    doi: 10.3799/dqkx.2023.029
    基金项目: 

    国家自然科学基金项目 41977238

    国家自然科学基金项目 51978634

    河南省自然科学基金青年基金项目 232300421331

    河南省高等学校重点科研项目 23A440005

    详细信息
      作者简介:

      吴禄源(1989-),男,博士,讲师,研究方向为岩石力学及岩土工程.ORCID:0000-0001-8403-9268. E-mail:wulymp@henu.edu.cn

      通讯作者:

      张建伟, E-mail:zjw101_0@163.com

    • 中图分类号: P64

    Prediction for Rock Compressive Strength Based on Ensemble Learning and Bayesian Optimization

    • 摘要: 岩石抗压强度是评估岩体工程稳定性的重要力学参数,传统统计回归方法对于岩石抗压强度预测存在一定的局限性.为此,提出了一种利用简单岩石力学参数实现岩石抗压强度智能预测的方法,首先收集了620组含不同类型岩石的三轴试验数据,然后分别采用随机森林(Random Forest,RF)、极限梯度提升树(XGBoost,XGB)和轻量梯度提升机(LightGBM,LGB)3种主流的集成学习算法建立了岩石抗压强度预测模型,使用贝叶斯优化算法在模型训练过程中进行超参数优化,最后利用决定系数(R2)、平均绝对百分比误差(MAPE)和均方根误差(RMSE)对优化后模型的泛化能力进行了综合评估和对比分析.此外,利用LGB模型对输入特征进行重要性分析,以评估不同输入特征对模型泛化性能的影响重要程度.研究结果表明:所建立的3种模型对岩石抗压强度均取得了较好的预测结果,其中LGB模型泛化性能优于另外两种模型(R2=0.978,RMSE=5.58,MAPE=9.70%),且运行耗时相对最少.弹性模量(E)、围压(σ3)和密度(ρ)对模型的泛化性能影响较大,泊松比(v)影响较小.提出的预测模型对于岩石抗压强度预测有良好的适用性,为机器学习与岩土工程的结合提供了新的思路.

       

    • 图  1  随机森林算法原理示意

      Fig.  1.  Schematic diagram of random forest algorithm

      图  2  不同模型下的预测值与实测值对比

      Fig.  2.  Comparison between predicted values and measured values under different models

      图  3  特征重要性分析结果

      Fig.  3.  Results of feature importance analysis

      表  1  基于贝叶斯优化后的3种模型最优超参数和运行耗时

      Table  1.   Optimal super parameters and running time of three models based on Bayesian optimization

      模型 超参数名称 最优取值 运行耗时(s)
      Random Forest n_estimators 915 13.625
      max_depth 13
      max_features 7
      min_samples_split 2
      min_samples_leaf 1
      LightGBM n_estimators 615 5.383
      max_depth 17
      gamma 26.251
      lambda 0.849
      alpha 24.513
      num_leaves 82
      learning_rate 0.201
      XGBoost n_estimators 985 24.558
      max_depth 3
      gamma 25.802
      lambda 2.881
      alpha 8.830
      learning_rate 0.185
      下载: 导出CSV

      表  2  预测模型评价指标

      Table  2.   Evaluation index of prediction model

      评价指标 计算公式 评判标准
      决定系数 $ {R}^{2}=1-\frac{{\sum\limits _{i=1}^{n}\left({y}_{i}-{\widehat{y}}_{i}\right)}^{2}}{{\sum \limits_{i=1}^{n}\left({y}_{i}-{\stackrel{-}{y}}_{i}\right)}^{2}} $ R2数值越大,模型性能越好
      均方根误差 $ \mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=\sqrt{\frac{1}{n}\times {\sum \limits_{i=1}^{n}\left({y}_{i}-{\widehat{y}}_{i}\right)}^{2}} $ RMSE数值越小,模型性能越好
      平均绝对百分比误差 $ \mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}=\sum\limits _{i=1}^{n}\left|\frac{{y}_{i}-{\widehat{y}}_{i}}{{y}_{i}}\right|\times \frac{100\mathrm{\%}}{n} $ MAPE数值越小,模型性能越好
      注:式中,n为预测样本的数量,$ {\stackrel{-}{y}}_{i} $和ŷi分别表示实测值yi的平均值和预测值.
      下载: 导出CSV
    • Aladejare, A. E., 2020. Evaluation of Empirical Estimation of Uniaxial Compressive Strength of Rock Using Measurements from Index and Physical Tests. Journal of Rock Mechanics and Geotechnical Engineering, 12(2): 256-268. https://doi.org/10.1016/j.jrmge.2019.08.001
      Arjmandpour, J., Hosseinitoudeshki, V., 2013. Estimation of Tensile Strength of Limestone from Some of Its Physical Properties via Simple Regression. Journal of Novel Applied Sciences, 2: 1041-1044.
      Bieniawski, Z. T., 1974. Estimating the Strength of Rock Materials. Journal of the South African Institute of Mining and Metallurgy, 74(8): 312-320. http://saimm.org.za/Journal/v074n08p312.pdf
      Breiman, L., 2001. Random Forests. Machine Learning, 45: 5-32. doi: 10.1023/A:1010933404324
      Cargill, J. S., Shakoor, A., 1990. Evaluation of Empirical Methods for Measuring the Uniaxial Compressive Strength of Rock. International Journal of Rock Mechanics and Mining Sciences, 27(6): 495-503. doi:https://doi.org/ 10.1016/0148-9062(90)91001-N
      Chen, T., Guestrin, C., 2016. XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, 785-794. https://doi.org/10.1145/2939672.2939785
      Çobanoğlu, İ., Çelik, S. B., 2008. Estimation of Uniaxial Compressive Strength from Point Load Strength, Schmidt Hardness and P-Wave Velocity. Bulletin of Engineering Geology and the Environment, 67: 491-498. doi: 10.1007/s10064-008-0158-x
      Cui, J. X., Yang, B., 2018. Survey on Bayesian Optimization Methodology and Applications. Journal of Software, 29(10): 3068-3090 (in Chinese with English abstract). http://www.researchgate.net/publication/330103291_Survey_on_Bayesian_Optimization_Methodology_and_Applications
      Culshaw, M. G., 2015. The ISRM Suggested Methods for Rock Characterization, Testing and Monitoring: 2007‒2014. Bulletin of Engineering Geology and the Environment, 74: 1499-1500. https://doi.org/10.1007/978-3-319-007713-0
      Edelbro, C., 2003. Rock Mass Strength: A Review. Department of Civil Engineering Division of Rock Mechanics, Beijing.
      Gokceoglu, C., 2002. A Fuzzy Triangular Chart to Predict the Uniaxial Compressive Strength of the Ankara Agglomerates from Their Petrographic Composition. Engineering Geology, 66: 39-51. https://doi.org/10.1016/S0013-7952(02)00023-6
      Goudie, A. S., 2006. The Schmidt Hammer in Geomorphological Research. Progress in Physical Geography, 30: 703-718. doi:https://doi.org/ 10.1177/0309133306071954
      Grima, M. A., Babuška, R., 1999. Fuzzy Model for the Prediction of Unconfined Compressive Strength of Rock Samples. International Journal of Rock Mechanics and Mining Sciences, 36: 339-349. doi:https://doi.org/ 10.1016/S0148-9062(99)00007-8
      Guo, Z. Z., Yin, K. L., Fu, S., et al., 2019. Evaluation of Landslide Susceptibility Based on GIS and WOE-BP Model. Earth Science, 44(12): 4299-4312 (in Chinese with English abstract). http://www.researchgate.net/publication/324390254_Evaluation_of_Landslide_Susceptibility_Based_on_GIS_and_WOE-BP_Model
      He, M., 2019. Deep Convolutional Neural Network for Fast Determination of the Rock Strength Parameters Using Drilling Data. International Journal of Rock Mechanics and Mining Sciences, 123: 104084. https://doi.org/10.1016/j.ijrmms.2019.104084
      Huang, F. M., Cao, Y., Fan, X. M., et al., 2021. Effects of Different Landslide Boundaries and Their Spatial Shapes on the Uncertainty of Landslide Susceptibility Prediction. Chinese Journal of Rock Mechanics and Engineering, 40(S02): 3227-3240 (in Chinese with English abstract).
      Huang, F. M., Chen, B., Mao, D. X., et al., 2023. Landslide Susceptibility Prediction Modeling and Interpretability Based on Self-Screening Deep Learning Model. Earth Science, 48(5): 1696-1710 (in Chinese with English abstract).
      Huang, X. H., Li, Z. H., Deng, T., et al., 2022. Uranium Potential Evaluation of the Zhuguangshan Granitic Pluton in South China Based on Machine Learning. Earth Science, 1-23 (in Chinese with English abstract).
      Jahed Armaghani, D., 2016. Application of Several Non- Linear Prediction Tools for Estimating Uniaxial Compressive Strength of Granitic Rocks and Comparison of Their Performances. Engineering with Computers, 32: 189-206. doi: 10.1007/s00366-015-0410-5
      Ke, G., 2017. Light GBM: A Highly Efficient Gradient Boosting Decision Tree. Advances in Neural Information Processing Systems, New Orleans, 30.
      Li, S., Chen, J., Liu, C., et al., 2021. Mineral Prospectivity Prediction via Convolutional Neural Networks Based on Geological Big Data. Journal of Earth Science, 32(2): 327-347. https://doi.org/10.1007/s12583-020-1365-z
      Li, W., Tan, Z. Y., 2016. Comparison on Rock Strength Prediction Models Based on MLR and LS-SVM. Mining Research and Development, 36(11): 36-40 (in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-KYYK201611008.htm
      Li, W. B., Fan, X. M., Huang, F. M., et al., 2021. Uncertainties of Landslide Susceptibility Modeling under Different Environmental Factor Connections and Prediction Models. Earth Science, 46(10): 3777-3795 (in Chinese with English abstract). http://www.sciencedirect.com/science/article/pii/S0341816221001090
      Li, Y. R., Zhang, Y. L., Wang, J. C., 2022. Survey on Bayesian Optimization Methods for Hyper-Parameter Tuning. Computer Science, 49(S01): 86-92 (in Chinese with English abstract).
      Mahmoodzadeh, A., 2022. Machine Learning Techniques to Predict Rock Strength Parameters. Rock Mechanics and Rock Engineering, 55: 1721-1741. doi: 10.1007/s00603-021-02747-x
      Miah, M. I., 2020. Machine Learning Approach to Model Rock Strength: Prediction and Variable Selection with Aid of Log Data. Rock Mechanics and Rock Engineering, 53: 4691-4715. doi: 10.1007/s00603-020-02184-2
      Mohamad, E. T., 2018. Rock Strength Estimation: A PSO-Based BP Approach. Neural Computing and Applications, 30: 1635-1646. doi: 10.1007/s00521-016-2728-3
      Sagi, O., Rokach, L., 2018. Ensemble Learning: A Survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 8: e1249. https://doi.org/10.1002/widm.1249
      Sarkar, K., 2010. Estimation of Strength Parameters of Rock Using Artificial Neural Networks. Bulletin of Engineering Geology and the Environment, 69: 599-606. doi: 10.1007/s10064-010-0301-3
      Singh, T., 2012. Correlation between Point Load Index and Uniaxial Compressive Strength for Different Rock Types. Rock Mechanics and Rock Engineering, 45: 259-264. doi: 10.1007/s00603-011-0192-z
      Tang, Z. L., Xu, Q. J., 2020. Rockburst Prediction Based on Nine Machine Learning Algorithms. Chinese Journal of Rock Mechanics and Engineering, 39(4): 773-781 (in Chinese with English abstract).
      Wang, M., Wan, W., 2019. A New Empirical Formula for Evaluating Uniaxial Compressive Strength Using the Schmidt Hammer Test. International Journal of Rock Mechanics and Mining Sciences, 123: 104094. https://doi.org/10.1016/j.ijrmms.2019.104094
      Wang, R., 2020. Application of Ultrasonic-Rebound Method in Fast Prediction of Rock Strength. Geotechnical and Geological Engineering, 38: 5915-5924. doi: 10.1007/s10706-020-01402-6
      Yang, K., Yuan, L., Qi, L. G., et al., 2013. Establishing Predictive Model for Rock Uniaxial Compressive Strength of No. 11-2 Coal Seam Roof in Huainan Mining Area. Chinese Journal of Rock Mechanics and Engineering, 32(10): 1991-1998 (in Chinese with English abstract). http://www.researchgate.net/publication/287704654_Establishing_predictive_model_for_rock_uniaxial_compressive_strength_of_No11-2coal_seam_roof_in_Huainan_mining_area
      Zhang, C. L., Zhang, C. P., Xu, J., 2015. Comparison Test of Rock Point Load Strength and Uniaxial Compressive Strength. Chinese Journal of Underground Space and Engineering, 11(S2): 447-451 (in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTotal-BASE2015S2014.htm
      Zhang, W. G., He, Y. W., Wang, L. Q., et al., 2023. Machine Learning Solution for Landslide Susceptibility Based on Hydrographic Division: Case Study of Fengjie County in Chongqing. Earth Science, 48(5): 2024-2038 (in Chinese with English abstract).
      Zhang, W. G., Li, H. R., Wu, C. Z., et al., 2021a. Stability Assessment of Underground Entry-Type Excavations Using Data-Driven RF and KNN Methods. Journal of Hunan University (Natural Sciences), 48(3): 164-172 (in Chinese with English abstract).
      Zhang, W. G., Tang, L. B., Chen, F. Y., et al., 2021b. Prediction for TBM Penetration Rate Using Four Hyperparameter Optimization Methods and Random Forest Model. Journal of Basic Science and Engineering, 29(5): 1186-1200 (in Chinese with English abstract). doi: 10.1007/978-981-16-6835-7_8
      Zhou, Z. H., 2016. Machine Learning. Tsinghua University Press, Beijing, 173(in Chinese).
      Zhou, Z. H., 2021. Ensemble Learning, Machine Learning. Springer, Berlin, 181-210.
      崔佳旭, 杨博, 2018. 贝叶斯优化方法和应用综述. 软件学报, 29(10): 3068-3090. https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201810011.htm
      郭子正, 殷坤龙, 付圣, 等, 2019. 基于GIS与WOE-BP模型的滑坡易发性评价. 地球科学, 44(12): 4299-4312. doi: 10.3799/dqkx.2018.555
      黄发明, 曹昱, 范宣梅, 等, 2021. 不同滑坡边界及其空间形状对滑坡易发性预测不确定性的影响规律. 岩石力学与工程学报, 40(S02): 3227-3240. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX2021S2023.htm
      黄发明, 陈彬, 毛达雄, 等, 2023. 基于自筛选深度学习的滑坡易发性预测建模及其可解释性. 地球科学, 48(5): 1696-1710. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202305003.htm
      黄鑫怀, 李增华, 邓腾, 等, 2022. 基于机器学习的华南诸广山花岗岩体铀矿潜力评价. 地球科学, 1-23.
      李文, 谭卓英, 2016. 基于MLR与LS-SVM的岩石强度预测模型比较. 矿业研究与开发, 36(11): 36-40. https://www.cnki.com.cn/Article/CJFDTOTAL-KYYK201611008.htm
      李文彬, 范宣梅, 黄发明, 等, 2021. 不同环境因子联接和预测模型的滑坡易发性建模不确定性. 地球科学, 46(10): 3777-3795. doi: 10.3799/dqkx.2021.042
      李亚茹, 张宇来, 王佳晨, 2022. 面向超参数估计的贝叶斯优化方法综述. 计算机科学, 49(S01): 86-92. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA2022S1013.htm
      汤志立, 徐千军, 2020. 基于9种机器学习算法的岩爆预测研究. 岩石力学与工程学报, 39(4): 773-781. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX202004011.htm
      杨科, 袁亮, 祁连光, 等, 2013. 淮南矿区11-2煤顶板岩石单轴抗压强度预测模型构建. 岩石力学与工程学报, 32(10): 1991-1998. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201310005.htm
      张春玲, 张传鹏, 徐静, 2015. 岩石点荷载强度与单轴抗压强度的对比试验. 地下空间与工程学报, 11(S2): 447-451. https://www.cnki.com.cn/Article/CJFDTOTAL-BASE2015S2014.htm
      仉文岗, 何昱苇, 王鲁琦, 等, 2023. 基于水系分区的滑坡易发性机器学习分析方法——以重庆市奉节县为例. 地球科学: 48(5): 2024-2038. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202305028.htm
      仉文岗, 李红蕊, 巫崇智, 等, 2021a. 基于RF和KNN的地下采场开挖稳定性评估. 湖南大学学报(自然科学版), 48(3): 164-172. https://www.cnki.com.cn/Article/CJFDTOTAL-HNDX202103017.htm
      仉文岗, 唐理斌, 陈福勇, 等, 2021b. 基于4种超参数优化算法及随机森林模型预测TBM掘进速度. 应用基础与工程科学学报, 29(5): 1186-1200. https://www.cnki.com.cn/Article/CJFDTOTAL-YJGX202105009.htm
      周志华, 2016. 机器学习. 北京: 清华大学出版社, 173.
    • 加载中
    图(3) / 表(2)
    计量
    • 文章访问数:  1487
    • HTML全文浏览量:  734
    • PDF下载量:  148
    • 被引次数: 0
    出版历程
    • 收稿日期:  2022-09-22
    • 网络出版日期:  2023-06-06
    • 刊出日期:  2023-05-25

    目录

      /

      返回文章
      返回