Machine learning method for CPTu based 3D stratification of New Zealand geotechnical database sites

Abstract Three-dimensional (3D) geotechnical site stratification is of vital importance in geotechnical practice. In this study, a set of methods for 3D site stratification based on CPTu measurements of New Zealand Geotechnical Database (NZGD) sites is proposed. One-dimensional (1D) soil stratification at discrete CPTu points is first conducted and then interpolated in 3D to achieve 3D site stratification. 1D soil stratification is achieved through a proposed soil classification model combined with a proposed soil layer boundary identification method, which achieves a correct soil profile length identification rate of 93%. The soil classification machine learning model classifies the soil within NZGD into three types, i.e. Gravel, Sand, and Silt, and is able to reflect the fines content for silty sand. The model innovatively incorporates local variation information of CPTu curves in the input for a random forest algorithm to significantly improve identification accuracy to over 90%. Accurately locating soil layer boundaries is achieved through proposing a modified WTMM boundary identification method. 3D site stratification is then realized through 3D interpolation of 1D stratification at discrete CPTu points using a generalized regression neural network (GRNN) method. The 3D site stratification method is validated for two independent geotechnical sites within NZGD, exhibiting the effectiveness of the proposed set of methods.

[1]  M. Jefferies,et al.  Use of CPTu to Estimate Equivalent SPT N60 , 1993 .

[2]  K. Cetin,et al.  CPT-Based Probabilistic Soil Characterization and Classification , 2009 .

[3]  Kok-Kwang Phoon,et al.  Geostatistical analysis of cone penetration test (CPT) sounding using the modified Bartlett test , 2004 .

[4]  G. Baecher.  Reliability and Statistics in Geotechnical Engineering , 2003 .

[5]  Dian-Qing Li,et al.  Bayesian identification of soil stratigraphy based on soil behaviour type index , 2019, Canadian Geotechnical Journal.

[6]  Peter K. Robertson,et al.  Cone-penetration testing in geotechnical practice , 1997 .

[7]  M. Ahmadi,et al.  Thin-layer effects on the CPT qc measurement , 2005 .

[8]  Robert Y. Liang,et al.  Quantifying stratigraphic uncertainties by stochastic simulation techniques based on Markov random field , 2016 .

[9]  Z. Zhang,et al.  Statistical to Fuzzy Approach toward CPT Soil Classification , 2000 .

[10]  Kok-Kwang Phoon,et al.  Identification of statistically homogeneous soil layers using modified Bartlett statistics , 2003 .

[11]  P. Robertson.  Interpretation of cone penetration tests — a unified approach , 2009 .

[12]  Chuanrong Zhang,et al.  Two-dimensional Markov chain simulation of soil type spatial distribution , 2004 .

[13]  R. S. Olsen,et al.  Soil Classification Using Electric Cone Penetrometer , 1981 .

[14]  S. Carle.  Use of a Transition Probability/Markov Approach to Improve Geostatistical of Facies Architecture , 2000 .

[15]  Kenneth Gavin,et al.  Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks , 2018, Adv. Eng. Informatics.

[16]  Jiro Kuwano,et al.  A kriging method of interpolation used to map liquefaction potential over alluvial ground , 2013 .

[17]  Peter K. Robertson,et al.  Estimating liquefaction induced ground settlements from CPT for level ground , 2002 .

[18]  Yali Amit,et al.  Shape Quantization and Recognition with Randomized Trees , 1997, Neural Computation.

[19]  Qianbing Zhang,et al.  BIM, machine learning and computer vision techniques in underground construction: Current status and future perspectives , 2020 .

[20]  Ping-Sien Lin,et al.  Simplified cone penetration test-based method for evaluating liquefaction resistance of soils , 2003 .

[21]  Yu Wang,et al.  Interpolation and stratification of multilayer soil property profile from sparse measurements using machine learning methods , 2020 .

[22]  P. Robertson.  Cone penetration test (CPT)-based soil behaviour type (SBT) classification system — an update , 2016 .

[23]  Kok-Kwang Phoon,et al.  Application of the Kriging-Based Response Surface Method to the System Reliability of Soil Slopes , 2013 .

[24]  C. Jang,et al.  Spatial estimation of the thickness of low permeability topsoil materials by using a combined ordinary-indicator kriging approach with multiple thresholds , 2016 .

[25]  M. Randolph,et al.  Comparing CPTU Q – F and Q – Δ u 2 /σ v0 ′soil classification charts , 2012 .

[26]  Milton E. Harr,et al.  Reliability-Based Design in Civil Engineering , 1987 .

[27]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[28]  Guo-Liang Chen,et al.  Three-dimensional site characterization with borehole data – A case study of Suzhou area , 2018 .

[29]  Yu Wang,et al.  Probabilistic assessment of liquefiable soil thickness considering spatial variability and model and parameter uncertainties , 2017 .

[30]  Abolfazl Eslami,et al.  Pile capacity by direct CPT and CPTu methods applied to 102 case histories , 1997 .

[31]  Jianye Ching,et al.  Cone penetration test (CPT)-based stratigraphic profiling using the wavelet transform modulus maxima method , 2015 .

[32]  Du Guang-yin.  Soil classification using CPTU data based upon cluster analysis theory , 2009 .

[33]  E. Martin,et al.  Assessment of the efficacies of correction procedures for multiple thin layer effects on Cone Penetration Tests , 2021 .

[34]  Xueyou Li,et al.  Simulating realistic geological stratigraphy using direction-dependent coupled Markov chain model , 2019, Computers and Geotechnics.

[35]  Yu Wang,et al.  Bayesian approach for probabilistic characterization of sand friction angles , 2010 .

[36]  Zhongjie Zhang,et al.  From Theory to Implementation of a CPT-Based Probabilistic and Fuzzy Soil Classification , 2008 .

[37]  Paul W. Mayne,et al.  Stratigraphic delineation by three-dimensional clustering of piezocone data , 2007 .

[38]  Mark Randolph,et al.  Bearing capacity on sand overlying clay soils: experimental and finite-element investigation of potential punch-through failure , 2013 .

[39]  Paolo Gardoni,et al.  Probabilistic soil identification based on cone penetration tests , 2008 .

[40]  C. H. Juang,et al.  Three-dimensional site characterisation: neural network approach , 2001 .

[41]  George P. Kouretzis,et al.  Bearing capacity of strip footings on sand over clay , 2019, Canadian Geotechnical Journal.

[42]  P. Robertson,et al.  Evaluating cyclic liquefaction potential using the cone penetration test , 1998 .

[43]  N. Ramsey.  A Calibrated Model For The Interpretation Of Cone Penetration Tests (Cpts) In North Sea Quaternary Soils , 2002 .

[44]  Michel Dekking,et al.  A Markov Chain Model for Subsurface Characterization: Theory and Applications , 2001 .

[45]  P. Robertson.  Soil classification using the cone penetration test , 1990 .

[46]  Yu Wang,et al.  Probabilistic identification of underground soil stratification using cone penetration tests , 2013 .

[47]  J. DeJong,et al.  Inverse filtering procedure to correct cone penetration data for thin-layer and transition effects , 2018 .

[48]  Mark Randolph,et al.  Analysis of Factors Influencing Soil Classification Using Normalized Piezocone Tip Resistance and Pore Pressure Parameters , 2008 .

[49]  P. Kurup,et al.  Prediction of Soil Composition from CPT Data using General Regression Neural Network , 2006 .

[50]  Paul W. Mayne,et al.  Objective Site Characterization Using Clustering of Piezocone Data , 2002 .

[51]  Yu Wang,et al.  Bayesian Approach for Probabilistic Site Characterization Using Cone Penetration Tests , 2013 .

[52]  Mark Cassidy,et al.  Probabilistic identification of soil stratification , 2016 .