Estimation of missing borehole data from physically-bounded multi-task Bayesian compressive sensing
ZHAO Tengyuan1, CHEN Jiuming1, LI Wei2, LI Wei3, WANG Dong4, XU Ling1
(1. School of Human Settlements and Civil Engineering, Xi?an Jiaotong University, Xi?an, Shaanxi 710043, China; 2. Faculty of Engineering, China University of Geosciences, Wuhan, Hubei 430074, China; 3. Shaanxi Construction Railway Construction Engineering Co., Ltd., Xi?an, Shaanxi 710016, China; 4. China Railway First Survey and Design Institute Group Co., Ltd., Xi?an,
Shaanxi 710043, China)
Abstract：The Cone Penetration Test (CPT) is one of the most important in-situ methods for geotechnical site investigation. To address issues encountered in engineering practice, such as localized data gaps and significant variations in CPT measurements due to probe malfunctions and abrupt changes in soil properties at stratigraphic interfaces, this paper proposes a physics-Bounded Multi-Task Bayesian Compressive Sensing (B-MTBCS) methodology. This approach establishes an information fusion mechanism for multiple CPT soundings within a Bayesian framework. By incorporating physical constraints—specifically, the non-negativity of CPT responses—as boundary conditions for predicted data, it allows for accurate estimation of missing CPT measurements. Validation through numerical simulations and real-world engineering case studies demonstrates the method’s effectiveness in recovering missing CPT data under complex geological conditions. Compared to the conventional Multi-Task Bayesian Compressive Sensing (MTBCS) approach, the proposed method reduces prediction errors by over 34%. These research findings provide substantial support for critical geotechnical applications, including parameter inversion, foundation bearing capacity evaluation, and liquefaction potential assessment in sandy soils.
[1] 郑 栋，李典庆，黄劲松. 基于 CPTU 和 MASW 勘察信息融合的二维土性参数剖面贝叶斯表征方法[J]. 应用基础与工程科学学报，2021，29(2)：337–354.(ZHENG Dong，LI Dianqing，HUANG Jingsong. A Bayesian characterization approach for 2D profiles of soil properties via integrating information from CPTU and MASW in site investigation[J]. Journal of Basic Science and Engineering，2021，29(2)：337–354.(in Chinese))
[2] 沈小克，蔡正银，蔡国军. 原位测试技术与工程勘察应用[J]. 土木工程学报，2016，49(2)：98–120.(SHEN Xiaoke，CAI Zhengyin，CAI Guojun. Applications of in-situ tests in site characterization and evaluation[J]. China Civil Engineering Journal，2016，49(2)：98–120.(in Chinese))
[3] 张继周，缪林昌，王华敬. 土性参数不确定性描述方法的探讨[J]. 岩土工程学报，2009，31(12)：1 936–1 940.(ZHANG Jizhou，MIAO Linchang，WANG Huajing. Methods for characterizing variability of soil parameters[J]. Chinese Journal of Geotechnical Engineering，2009，31(12)：1 936–1 940.(in Chinese))
[4] CAO Z J，ZHENG S，LI D Q，et al. Bayesian identification of soil stratigraphy based on soil behaviour type index[J]. Canadian Geotechnical Journal，2019，56(4)：570–586.
[5] 胡 越，王 宇. 静力触探识别场地土层分布的贝叶斯学习方法研究[J]. 工程地质学报，2020，28(5)：966–972.(HU Yue，WANG Yu. Identification of subsurface soil stratification using cone penetration tests and Bayesian learning[J]. Journal of Engineering Geology，2020，28(5)：966–972.(in Chinese))
[6] BLOOMBERG D，UPCHURCH S B，HAYDEN M L，et al. Cone-penetrometer exploration of Sinkholes：Stratigraphy and soil properties[J]. Environmental Geology and Water Sciences，1988，12(2)：99–105.
[7] 曹子君，胡 超，王亚飞，等. 基于静力触探试验和变维联合后验分布的土层剖面高效优化识别方法[J]. 岩土工程学报，2025，47(2)：346–354.(CAO Zijun，HU Chao，WANG Yafei，et al. Efficient optimization identification method for soil stratification based on cone penetration test and joint posterior distribution of variable dimensionality[J]. Chinese Journal of Geotechnical Engineering，2025，47(2)：346–354.(in Chinese))
[8] 曹子君，郑 硕，李典庆，等. 基于静力触探的土层自动划分方法与不确定性表征[J]. 岩土工程学报，2018，40(2)：336–345.(CAO Zijun，ZHENG Shuo，LI Dianqing，et al. Probabilistic characterization of underground stratigraphy and its uncertainty based on cone penetration test[J]. Chinese Journal of Geotechnical Engineering，2018，40(2)：336–345.(in Chinese))
[9] WANG Y，FU C，HUANG K. Probabilistic assessment of liquefiable soil thickness considering spatial variability and model and parameter uncertainties[J]. Géotechnique，2017，67(3)：228–241.
[10] 赵泽宁，段 伟，蔡国军，等. 中国规范CPT砂土液化判别模型的改进与概率形式[J]. 岩土工程学报，2025，47(4)：869–876. (ZHAO Zening，DUAN Wei，CAI Guojun，et al. Improvement and probabilistic form of CPT-based liquefaction evaluation model in Chinese code[J]. Chinese Journal of Geotechnical Engineering，2025，47(4)：869–876.(in Chinese))
[11] HAKAM A，WARZUQNI S，ADJI B M，et al. CPT-DR-D50 correlations of sands for liquefaction potential analysis[C]// E3S Web of Conferences. [S. l.]：[s. n.]，2020：02017.
[12] 段 伟，赵泽宁，蔡国军，等. 基于CPT土分类图及其指标的土体液化评价模型[J]. 岩土工程学报，2025，待刊.(DUAN Wei，ZHAO Zening，CAI Guojun，et al. Soil liquefaction evaluation model based on soil behavior type classification chart and its indices from cone penetration testing[J]. Chinese Journal of Geotechnical Engineering, 2025，to be Pressed.(in Chinese))
[13] 张小丁，廖亚楠，陈 涛. 静力触探试验在岩土工程勘察中的应用[J]. 中国金属通报，2019，(12)：213–214.(ZHANG Xiaoding，LIAO Yanan，CHEN Tao. Application of cone penetration test in geotechnical investigation[J]. China Metal Bulletin，2019，(12)：213–214. (in Chinese))
[14] HEIDARIE G S，ESLAMI A，JAMSHIDI C R. Probabilistic assessment of model uncertainty for prediction of pile foundation bearing capacity; static analysis，SPT and CPT-based methods[J]. Geotechnical and Geological Engineering，2020，38(5)：5 023–5 041.
[15] 宋 跃，顾晓强，胡 靖，等. 砂土静力触探试验的 DEM-FDM 耦合数值模拟研究[J]. 岩土工程学报，2024，47(6)：1 249–12 58.(SONG Yue，GU Xiaoqiang，HU Jing，et al. DEM-FDM coupled simulation of cone penetration tests in a virtual calibration chamber with sand[J]. Chinese Journal of Geotechnical Engineering，2024，47(6)：1 249– 1 258.(in Chinese)
[16] CHING J Y，YANG Z Y，PHOON K K. Dealing with nonlattice data in three-dimensional probabilistic site characterization[J]. Journal of Engineering Mechanics，2021，147(5)：06021003.
[17] YANG Z Y，CHING J Y. Simulation of three-dimensional random field conditioning on incomplete site data[J]. Engineering Geology，2021，281：105987.
[18] ZHAO T Y，WANG Y，LU S F，et al. Fast stratification of geological cross-section from CPT results with missing data using multitask and modified Bayesian compressive sensing[J]. Canadian Geotechnical Journal，2023，60(12)：1 812–1 834.
[19] ZHANG W C，SHI P X，ZHAO H J，et al. Domain knowledge-guided Bayesian evolutionary trees for estimating the compression modulus of soils containing missing values[J]. Engineering Applications of Artificial Intelligence，2024，133：108356.
[20] CAO E H，BAO T F，LIU Y T，et al. A data enhancement-based quadratic imputation framework for consecutive missing values considering spatiotemporal characteristics of dam deformation[J]. Journal of Civil Structural Health Monitoring，2024，14(2)：431–447.
[21] XU B，RONG Z，PANG R，et al. A novel method for settlement imputation and monitoring of earth-rockfill dams subjected to large-scale missing data[J]. Advanced Engineering Informatics.，2024，62：102642.
[22] SHI N，LI Y L，WEN L F，et al. Rapid prediction of landslide dam stability considering the missing data using XGBoost algorithm[J]. Landslides，2022，19(12)：2 951–2 963.
[23] WANG Y，ZHAO T Y. Interpretation of soil property profile from limited measurement data：a compressive sampling perspective[J]. Canadian Geotechnical Journal，2016，53(9)：1 547–1 559.
[24] WANG Y，HU Y，ZHAO T Y. Cone penetration test(CPT)-based subsurface soil classification and zonation in two-dimensional vertical cross section using Bayesian compressive sampling[J]. Canadian Geotechnical Journal，2020，57(7)：947–958.
[25] 朱文清，赵腾远，宋 超，等. 基于吉布斯采样与压缩感知的二维非平稳CPT数据快速插值方法[J]. 土木与环境工程学报(中英文)，2022，44(5)：98–108.(ZHU Wenqing，ZHAO Tengyuan，SONG Chao，et al. Efficient interpolation method for 2D non-stationary CPT data using Gibbs sampling and compressive sampling[J]. Journal of Civil and Environmental Engineering，2022，44(5)：98–108.(in Chinese))
[26] ZHANG J，LI J P，ZHANG L M，et al. Calibrating cross-site variability for reliability-based design of pile foundations[J]. Computers and Geotechnics，2014，62：154–163.
[27] 许 领，王文龙，赵腾远. 考虑场地相似性的小样本边坡可靠度分析[J]. 岩石力学与工程学报，2025，44(4)：977–988.(XU Ling，WANG Wenlong，ZHAO Tengyuan. Reliability analysis of slopes from sparse measurements considering sites similarity[J]. Chinese Journal of Rock Mechanics and Engineering，2025，44(4)：977–988.(in Chinese))
[28] ANG A H S，TANG W H. Probability concepts in engineering: emphasis on applications to civil and environmental engineering[M]. 2nd ed. New York：Wiley，2007：346–371.
[29] SIVIA D S. Data analysis：A Bayesian tutorial[M]. [S. l.]：Oxford University Press，2006：30–90
[30] MOHASEL A H，DOMKE J. Reflection, refraction, and Hamiltonian Monte Carlo[C]// Proceedings of the 29th International Conference on Neural Information Processing Systems-Volume 2. Cambridge，MA，USA：MIT Press，2015：3 007–3 015.
[31] PAKMAN A，PANINSKI L. Exact Hamiltonian Monte Carlo for truncated multivariate gaussians[J]. Journal of Computational and Graphical Statistics，2014，23(2)：518–542.
[32] GAST D T. Dykes and Embankments：a geostatistical analysis of soft terrain[Ph. D. Thesis][D]. [S. l.]：Delft University of Technology，2020..
[33] GONG W P，ZHAO C，JUANG C H，et al. Stratigraphic uncertainty modelling with random field approach[J]. Computers and Geotechnics，2020，125：103681
2025, Vol. 44 
