Estimation of uniaxial compressive strength based on fully Bayesian Gaussian process regression and model selection

SONG Chao; ZHAO Tengyuan; XU Ling

doi:10.11779/CJGE20220734

Volume 45 Issue 8

Turn off MathJax

Article Contents

Abstract

References

Supplements

Chinese Journal of Geotechnical Engineering > 2023 > 45(8): 1664-1673. > DOI: 10.11779/CJGE20220734

SONG Chao, ZHAO Tengyuan, XU Ling. Estimation of uniaxial compressive strength based on fully Bayesian Gaussian process regression and model selection[J]. Chinese Journal of Geotechnical Engineering, 2023, 45(8): 1664-1673. DOI: 10.11779/CJGE20220734

Citation:

PDF (2744 KB)

Estimation of uniaxial compressive strength based on fully Bayesian Gaussian process regression and model selection

School of Human Settlements and Civil Engineering, Xi'an Jiaotong University, Xi'an 710049, China

More Information

Received Date: June 08, 2022
Available Online: February 26, 2023

Graphical Abstract

Abstract

Abstract

In order to establish an optimal model for estimating the uniaxial compressive strength (UCS) of rocks as well as its reasonable estimation, a fully Bayesian Gaussian process regression method (fB-GPR) is proposed by combining the Gaussian process regression (GPR), Bayesian framework and Markov Chain Monte Carlo (MCMC) simulation. The proposed fB-GPR approach is compared with different model selection methods, such as the Akaike information criterion (AIC), Bayesian information criterion (BIC), deviation information criterion (DIC), Kullback information criterion (KIC), etc. The results show that the proposed fB-GPR method performs better than other methods. In 100 random trials, the probability of M-7 being selected as the optimal model by fB-GPR method reaches 100%, and the accuracy of selecting the optimal model is far higher than other model selection methods. When the measurement noise reaches 50% of UCS standard deviation, the proposed fB-GPR can still achieve model selection accurately, which shows that the fB-GPR approach is robust and accurate, and is less affected by the measurement noise associated with UCS, comparing with other model selection methods. The proposed fB-GPR therefore provides a new way for establishing the optimal estimation model as well as reasonable estimation for the key geotechnical parameters in practice.

FullText(HTML)

References (25)

References

[1]	李文, 谭卓英. 基于P波模量的岩石单轴抗压强度预测[J]. 岩土力学, 2016, 37(增刊2): 381-387. doi: 10.16285/j.rsm.2016.S2.049 LI Wen, TAN Zhuoying. Prediction of uniaxial compressive strength of rock based on P-wave modulus[J]. Rock and Soil Mechanics, 2016, 37(S2): 381-387. (in Chinese) doi: 10.16285/j.rsm.2016.S2.049
[2]	孙欢, 刘晓丽, 王恩志, 等. 基于X射线断层扫描预测岩石单轴抗压强度[J]. 岩石力学与工程学报, 2019, 38(增刊2): 3575-3582. doi: 10.13722/j.cnki.jrme.2019.0614 SUN Huan, LIU Xiaoli, WANG Enzhi, et al. Prediction on uniaxial compression strength of rocks with X-ray computed tomography[J]. Chinese Journal of Rock Mechanics and Engineering, 2019, 38(S2): 3575-3582. (in Chinese) doi: 10.13722/j.cnki.jrme.2019.0614
[3]	沙鹏, 张庆同, 林军, 等. 基于点荷载强度的火成岩单轴抗压强度原位估算[J]. 岩土力学, 2020, 41(增刊2): 1-10. doi: 10.16285/j.rsm.2020.0371 SHA Peng, ZHANG Qingtong, LIN Jun, et al. In-situ estimation of uniaxial compressive strength of igneous rock based on point load strength[J]. Rock and Soil Mechanics, 2020, 41(S2): 1-10. (in Chinese) doi: 10.16285/j.rsm.2020.0371
[4]	ÇELIK S B. Prediction of uniaxial compressive strength of carbonate rocks from nondestructive tests using multivariate regression and LS-SVM methods[J]. Arabian Journal of Geosciences, 2019, 12(6): 1-17.
[5]	LIU D S, LIU H L, WU Y, et al. Characterization of geo-material parameters: gene concept and big data approach in geotechnical engineering[J]. Geosystems and Geoenvironment, 2022, 1(1): 100003. doi: 10.1016/j.geogeo.2021.09.003
[6]	HAN L, WANG L, DING X M, et al. Similarity quantification of soil parametric data and sites using confidence ellipses[J]. Geoscience Frontiers, 2022, 13(1): 101280. doi: 10.1016/j.gsf.2021.101280
[7]	MATIN S S, FARAHZADI L, MAKAREMI S, et al. Variable selection and prediction of uniaxial compressive strength and modulus of elasticity by random forest[J]. Applied Soft Computing, 2018, 70: 980-987. doi: 10.1016/j.asoc.2017.06.030
[8]	FATTAHI H. Applying soft computing methods to predict the uniaxial compressive strength of rocks from Schmidt hammer rebound values[J]. Computational Geosciences, 2017, 21(4): 665-681. doi: 10.1007/s10596-017-9642-3
[9]	FENG X, JIMENEZ R. Estimation of deformation modulus of rock masses based on Bayesian model selection and Bayesian updating approach[J]. Engineering Geology, 2015, 199: 19–27. doi: 10.1016/j.enggeo.2015.10.002
[10]	WU Z X, JI H, YU C, et al. EPR-RCGA-based modelling of compression index and RMSE-AIC-BIC-based model selection for Chinese marine clays and their engineering application[J]. Journal of Zhejiang University-SCIENCE A, 2018, 19(3): 211-224. [LinkOut] doi: 10.1631/jzus.A1700089
[11]	NG I T, YUEN K V, LAU C H. Predictive model for uniaxial compressive strength for Grade Ⅲ granitic rocks from Macao[J]. Engineering Geology, 2015, 199: 28–37. doi: 10.1016/j.enggeo.2015.10.008
[12]	RASMUSSEN C E, WILLIAMS C K I. Gaussian processes for machine learning[M]. Cambridge Mass: MIT Press, 2006.
[13]	苏国韶, 肖义龙. 边坡可靠度分析的高斯过程方法[J]. 岩土工程学报, 2011, 33(6): 916-920. http://www.cgejournal.com/cn/article/id/14030 SU Guoshao, XIAO Yilong. Gaussian process method for slope reliability analysis[J]. Chinese Journal of Geotechnical Engineering, 2011, 33(6): 916-920. (in Chinese) http://www.cgejournal.com/cn/article/id/14030
[14]	赵腾远, 宋超, 何欢. 小样本条件下江苏软土路基回弹模量的贝叶斯估计——基于静力触探数据与高斯过程回归的建模分析[J]. 岩土工程学报, 2021, 43(增刊2): 137–141. doi: 10.11779/CJGE2021S2033 ZHAO Tengyuan, SONG Chao, HE Huan. Bayesian estimation of resilient modulus of Jiangsu soft soils from sparse data—Gaussian process regression and cone penetration test data-based modelling and analysis[J]. Chinese Journal of Geotechnical Engineering, 2021, 43(S2): 137–141. (in Chinese) doi: 10.11779/CJGE2021S2033
[15]	蒋水华, 刘源, 张小波, 等. 有限数据条件下空间变异岩土力学参数随机反演分析及比较[J]. 岩石力学与工程学报, 2020, 39(6): 1265–1276. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX202006017.htm JIANG Shuihua, LIU Yuan, ZHANG Xiaobo, et al. Stochastic back analysis and comparison of spatially varying geotechnical mechanical parameters based on limited data[J]. Chinese Journal of Rock Mechanics and Engineering, 2020, 39(6): 1265–1276. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX202006017.htm
[16]	VANHATALO J, RIIHIMÄKI J, HARTIKAINEN J, et al. GPstuff: Bayesian modeling with Gaussian processes[J]. Journal of Machine Learning Research, 2013, 14(1): 1175-1179.
[17]	仉文岗, 顾鑫, 刘汉龙, 等. 基于贝叶斯更新的非饱和土坡参数概率反演及变形预测[J]. 岩土力学, 2022, 43(4): 1112–1122. https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX202204024.htm ZHANG Wengang, GU Xin, LIU Hanlong, et al. Probabilistic back analysis of soil parameters and displacement prediction of unsaturated slopes using Bayesian updating[J]. Rock and Soil Mechanics, 2022, 43(4): 1112-1122. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX202204024.htm
[18]	ZHAO T Y, SONG C, LU S F, et al. Prediction of uniaxial compressive strength using fully Bayesian Gaussian process regression (fB-GPR) with model class selection[J]. Rock Mechanics and Rock Engineering, 2022, 55(10): 6301-6319.
[19]	CZADO C. Analyzing dependent data with vine copulas[J]. Lecture Notes in Statistics, Springer, 2019.
[20]	付代光, 刘江平, 周黎明, 等. 基于贝叶斯理论的软夹层多模式瑞雷波频散曲线反演研究[J]. 岩土工程学报, 2015, 37(2): 321-329. doi: 10.11779/CJGE201502016 FU Daiguang, LIU Jiangping, ZHOU Liming, et al. Inversion of multimode Rayleigh-wave dispersion curves of soft interlayer based on Bayesian theory[J]. Chinese Journal of Geotechnical Engineering, 2015, 37(2): 321-329. (in Chinese) doi: 10.11779/CJGE201502016
[21]	唐小松, 李典庆, 周创兵, 等. 基于Bootstrap方法的岩土体参数联合分布模型识别[J]. 岩土力学, 2015, 36(4): 913-922. https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX201504001.htm TANG Xiaosong, LI Dianqing, ZHOU Chuangbing, et al. Bootstrap method for joint probability distribution identification of correlated geotechnical parameters[J]. Rock and Soil Mechanics, 2015, 36(4): 913-922. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YTLX201504001.htm
[22]	HURVICH C M, TSAI C L. Regression and time series model selection in small samples[J]. Biometrika, 1989, 76(2): 297-307.
[23]	SPIEGELHALTER D J, BEST N G, CARLIN B P, et al. Bayesian measures of model complexity and fit[J]. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2002, 64(4): 583-639.
[24]	CAVANAUGH J E. A large-sample model selection criterion based on Kullback's symmetric divergence[J]. Statistics & Probability Letters, 1999, 42(4): 333-343.
[25]	SEGHOUANE A K. Multivariate regression model selection from small samples using Kullback's symmetric divergence[J]. Signal Processing, 2006, 86(8): 2074-2084.