地质统计学变异函数贝叶斯模型选择方法与比较

张一凡; 张璐璐; 徐加宝

doi:10.19509/j.cnki.dzkq.tb20240202

您当前的位置：

首页 >

文章列表页 >

地质统计学变异函数贝叶斯模型选择方法与比较

更新时间：2026-11-05

- 地质统计学变异函数贝叶斯模型选择方法与比较
- Bulletin of Geological Science and Technology Vol. 44, Issue 2, Pages: 38-47(2025)
- 作者机构：
  
  上海交通大学海洋工程国家重点实验室
- 作者简介：
- 基金信息：
- DOI：10.19509/j.cnki.dzkq.tb20240202
  CLC： P628.2;TU413
- Online First：21 March 2025，
  
  Published：2025
- 稿件说明：
移动端阅览
张一凡, 张璐璐, 徐加宝. 地质统计学变异函数贝叶斯模型选择方法与比较[J]. Bulletin of Geological Science and Technology, 2025, 44(2): 38-47.
DOI：

张一凡, 张璐璐, 徐加宝. 地质统计学变异函数贝叶斯模型选择方法与比较[J]. Bulletin of Geological Science and Technology, 2025, 44(2): 38-47. DOI： 10.19509/j.cnki.dzkq.tb20240202.

摘要

变异函数量化了空间2点地质属性的变异性，对地质统计分析至关重要。当地质数据随空间坐标呈现趋势变化时，正确选择和估计变异函数十分困难。为实现变异函数的模型选择和参数估计，提出了基于贝叶斯理论的变异函数选择方法，采用拉普拉斯近似方法将后验概率分布近似为高斯分布。首先计算出参数的后验概率分布，随后分别计算每个备选变异函数的贝叶斯模型证据，以确定最优模型。探讨了3种模型选择方法在变异函数选择中的适用性，包括贝叶斯模型证据（BME）、Akaike information criterion(AIC)识别准则和Bayesian information criterion(BIC)识别准则。通过实测静力触探试验的锥端阻力数据，说明了该方法，并从模型拟合度和复杂度罚值2个方面比较3种方法在变异函数模型选择中的差异性。研究表明，给定试验数据条件下，BME能够合理地考虑变异函数的拟合度和复杂性；而AIC和BIC识别准则在模型参数个数相同时，仅能反映不同变异函数的拟合度差异，因此，在这种情况下推荐采用BME选择变异函数。本研究方法能够在考虑趋势项参数条件下合理地选择地质统计学变异函数，所选最优变异函数与试验变异函数较一致，为地质统计学分析提供了有效的参考。

Abstract

ObjectiveThe variogram quantifies the variability of geological attributes between two spatial points and is of crucial significance for geostatistical analysis. When geological data exhibit a trend variation along spatial coordinates

the accurate selection and estimation of the variogram become exceptionally difficult. MethodsTo realize the model selection and parameter estimation of the variogram

this paper presents a variogram selection approach based on Bayesian theory

employing the Laplace approximation method to approximate the posterior probability distribution as a Gaussian one. Firstly

the posterior probability distribution of the parameters is computed

and subsequently

the Bayesian model evidence (BME) of each alternative variogram is calculated respectively to determine the optimal model. This study investigates the applicability of three model selection methods in the selection of variograms

encompassing Bayesian model evidence (BME)

Akaike information criterion (AIC)

and Bayesian information criterion (BIC). ResultsThe proposed method is elucidated through the measured cone tip resistance data from static cone penetration tests

and the disparities among the three methods in the selection of variogram models are compared from the perspectives of model fitting and complexity penalty.ConclusionThe research reveals that

under the given experimental data conditions

BME can rationally take into account the fitting degree and complexity of the variogram; while the AIC and BIC identification criteria can merely reflect the fitting degree differences of different variograms when the number of model parameters is the same. Consequently

in such circumstances

BME is recommended for the selection of variograms. The method proposed in this study is capable of reasonably selecting the geostatistical variogram considering the trend term parameters

and the selected optimal variogram is relatively consistent with the experimental variogram

providing an effective reference for geostatistical analysis.

关键词

Keywords

references

Zhao Tengyuan,Song Chao,Lu Shifeng,Xu Ling.Prediction of Uniaxial Compressive Strength Using Fully Bayesian Gaussian Process Regression (fB-GPR) with Model Class Selection[J].Rock Mechanics and Rock Engineering,2022.

Jiabao Xu,Lulu Zhang,Jinhui Li,Zijun Cao,Haoqing Yang,Xiangyu Chen.Probabilistic estimation of variogram parameters of geotechnical properties with a trend based on Bayesian inference using Markov chain Monte Carlo simulation[J].Georisk Assessment and Management of Risk for Engineered Systems and Geohazards,2020.

Luo N,Arwade S R,DeGroot D J.Probabilistic analysis of offshore geotechnical site investigation in a homogeneous stiff clay deposit[J].Journal of Physics: Conference Series,2020.

Uzielli Marco,Mayne Paul W..Probabilistic assignment of effective friction angles of sands and silty sands FROM CPT using quantile regression[J].Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards,2019.

Qiushi Chen,Chaofeng Wang,C. Hsein Juang.Probabilistic and spatial assessment of liquefaction-induced settlements through multiscale random field models[J].Engineering Geology,2016.

Zijun Cao,Yu Wang,Dianqing Li.Quantification of prior knowledge in geotechnical site characterization[J].Engineering Geology,2016.

X. Y. Li,L. M. Zhang,J. H. Li.Using Conditioned Random Field to Characterize the Variability of Geologic Profiles[J].Journal of Geotechnical and Geoenvironmental Engineering,2015.

Zijun Cao,Yu Wang.Bayesian model comparison and selection of spatial correlation functions for soil parameters[J].Structural Safety,2014.

Zijun Cao,Yu Wang.Bayesian Model Comparison and Characterization of Undrained Shear Strength[J].Journal of Geotechnical and Geoenvironmental Engineering,2014.

Armin W. Stuedlein,Steven L. Kramer,Pedro Arduino,Robert D. Holtz.Geotechnical Characterization and Random Field Modeling of Desiccated Clay[J].Journal of Geotechnical and Geoenvironmental Engineering,2012.

Ka-Veng Yuen.Recent developments of Bayesian model class selection and applications in civil engineering[J].Structural Safety,2010.

Kok-Kwang Phoon,Ser-Tong Quek,Ping An.Identification of Statistically Homogeneous Soil Layers Using Modified Bartlett Statistics[J].Journal of Geotechnical and Geoenvironmental Engineering,2003.

Kok-Kwang Phoon,Fred H Kulhawy.Characterization of geotechnical variability[J].Canadian Geotechnical Journal,1999.

John T. Christian,Charles C. Ladd,Gregory B. Baecher.Reliability Applied to Slope Stability Analysis[J].Journal of Geotechnical Engineering,1994.

钟世英,刘长飞,时文浩.MATLAB工具箱自动识别探地雷达空洞图像技术[J].计算机辅助工程,2024(04).

张闻宇,张智刚,颜小挺,李杰浩.AI大模型在MATLAB课程教学中的应用[J].现代信息科技,2024(24).

杨文礼,白光顺,孙滨,张卫锋,杨雪梅.云南哀牢山地区地质灾害因子敏感性和易发性评价——以新平县为例[J].中国地质灾害与防治学报,2024(06).

邹鹤民.基于贝叶斯网络的公路隧道围岩失稳风险评估及系统开发[J].地质科技通报,2024(06).

朱悦,彭荣华,胡祥云,周文龙,黄顺聪,鲁杏.基于变维度贝叶斯反演的地热黏土盖层音频大地电磁探测能力研究[J].地质科技通报,2024(03).

李建文,赵文,吴振坤,徐小兵,王庆涛,段隆臣.煤矿采空区覆岩“三带”智能识别方法[J].煤田地质与勘探,2024(04).

揭鸿鹄,蒋水华,常志璐,黄劲松,黄发明.融合多源信息的降雨入渗边坡概率反分析及可靠度预测[J].中国地质灾害与防治学报,2024(01).

花卫华,宿紫莹,朱玉华,段剑超,张文,庞世龙,肖海清,刘修国.大范围地质体分块建模方法[J].地质科技通报,2023(06).

汪泉娟,孙敬锋,杨英杰,张将伟,王国梁,马恩泽,刘进,赵鑫.克里金方法与深度学习方法用于浅层地下水位估计的对比研究：以深汕特别合作区为例[J].地质科技通报,2024(04).

王继玲,周维博,孙梨梨,王奕勐.石川河富平地下水库渗透系数空间变异性研究[J].水文地质工程地质,2023(03).

郭秋麟,任洪佳,于京都,刘继丰,陈宁生.基于贝叶斯网络的油气勘探风险预测方法——以准噶尔盆地腹部侏罗系三工河组为例[J].中国石油勘探,2023(01).

黄宇,杨荣森,韩晓东,范建伟,倪春中.基于定向半变异函数的裂隙网络空间变异性分析[J].地质科技通报,2023(02).

宋超,赵腾远,许领.基于贝叶斯高斯过程回归与模型选择的岩石单轴抗压强度估计方法[J].岩土工程学报,2023(08).

杨灿,刘磊磊,张遗立,朱文卿,张绍和.基于贝叶斯优化机器学习超参数的滑坡易发性评价[J].地质科技通报,2022(02).

夏侯云山,张抒,唐辉明,刘晓,吴琼.考虑参数空间变异结构的结构化交叉约束随机场模拟方法研究[J].岩土力学,2019(12).

张国萍,王林,曹子君,李典庆.考虑不确定性的土水特征曲线模型确定方法比较研究[J].自然灾害学报,2018(04).

田密,李典庆,曹子君,方国光,王宇.基于贝叶斯理论的土性参数空间变异性量化方法[J].岩土力学,2017(11).

林军,蔡国军,邹海峰,刘松玉.基于随机场理论的江苏海相黏土空间变异性评价研究[J].岩土工程学报,2015(07).

左自波,张璐璐,程演,王建华,何晔.基于MCMC法的非饱和土渗流参数随机反分析[J].岩土力学,2013(08).

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

塔里木盆地低地温场成因与不同区带差异机制

基于文献计量学的地下水年龄研究进展与趋势分析

乌蒙山区特长极深隧道隧址区地应力场分布特征及反演分析

《地质科技通报》征稿启事

新疆西天山阿尔恰勒铅锌铜矿床磁铁矿、黑柱石成因及地质意义

Related Author

No data

Related Institution

No data

AI问答

⁰