Prediction of rockhead using a hybrid N-XGBoost machine learning framework

The spatial information of rockhead is crucial for the design and construction of tunneling or underground excavation. Although the conventional site investigation methods (i.e. borehole drilling) could provide local engineering geological information, the accurate prediction of the rockhead position with limited borehole data is still challenging due to its spatial variation and great uncertainties involved. With the development of computer science, machine learning (ML) has been proved to be a promising way to avoid subjective judgments by human beings and to establish complex relationships with mega data automatically. However, few studies have been reported on the adoption of ML models for the prediction of the rockhead position. In this paper, we proposed a robust probabilistic ML model for predicting the rockhead distribution using the spatial geographic information. The framework of the natural gradient boosting (NGBoost) algorithm combined with the extreme gradient boosting (XGBoost) is used as the basic learner. The XGBoost model was also compared with some other ML models such as the gradient boosting regression tree (GBRT), the light gradient boosting machine (LightGBM), the multivariate linear regression (MLR), the artificial neural network (ANN), and the support vector machine (SVM). The results demonstrate that the XGBoost algorithm, the core algorithm of the probabilistic N-XGBoost model, outperformed the other conventional ML models with a coefficient of determination (R²) of 0.89 and a root mean squared error (RMSE) of 5.8 m for the prediction of rockhead position based on limited borehole data. The probabilistic N-XGBoost model not only achieved a higher prediction accuracy, but also provided a predictive estimation of the uncertainty. Thus, the proposed N-XGBoost probabilistic model has the potential to be used as a reliable and effective ML algorithm for the prediction of rockhead position in rock and geotechnical engineering.     

Machine learning and data-driven prediction of pore pressure from geophysical logs: A case study for the Mangahewa gas field,New Zealand

Pore pressure is an essential parameter for establishing reservoir conditions, geological interpretation and drilling programs. Pore pressure prediction depends on information from various geophysical logs, seismic, and direct down-hole pressure measurements. However, a level of uncertainty accompanies the prediction of pore pressure because insufficient information is usually recorded in many wells. Applying machine learning (ML) algorithms can decrease the level of uncertainty of pore pressure prediction uncertainty in cases where available information is limited. In this research, several ML techniques are applied to predict pore pressure through the over-pressured Eocene reservoir section penetrated by four wells in the Mangahewa gas field, New Zealand. Their predictions substantially outperform, in terms of prediction performance, those generated using a multiple linear regression (MLR) model. The geophysical logs used as input variables are sonic, temperature and density logs, and some direct pore pressure measurements were available at the reservoir level to calibrate the predictions. A total of 25,935 data records involving six well-log input variables were evaluated across the four wells. All ML methods achieved credible levels of pore pressure prediction performance. The most accurate models for predicting pore pressure in individual wells on a supervised basis are decision tree (DT), adaboost (ADA), random forest (RF) and transparent open box (TOB). The DT achieved root mean square error (RMSE) ranging from 0.25 psi to 14.71 psi for the four wells. The trained models were less accurate when deployed on a semi-supervised basis to predict pore pressure in the other wellbores. For two wells (Mangahewa-03 and Mangahewa-06), semi-supervised prediction achieved acceptable prediction performance of RMSE of 130–140 psi; while for the other wells, semi-supervised prediction performance was reduced to RMSE > 300 psi. The results suggest that these models can be used to predict pore pressure in nearby locations, i.e. similar geology at corresponding depths within a field, but they become less reliable as the step-out distance increases and geological conditions change significantly. In comparison to other approaches to predict pore pressures, this study has identified that application of several ML algorithms involving a large number of data records can lead to more accurate prediction results.     

Rockhead profile simulation using an improved generation method of conditional random field

Rockhead profile is an important part of geological profiles and can have significant impacts on some geotechnical engineering practice, and thus, it is necessary to establish a useful method to reverse the rockhead profile using site investigation results. As a general method to reflect the spatial distribution of geo-material properties based on field measurements, the conditional random field (CRF) was improved in this paper to simulate rockhead profiles. Besides, in geotechnical engineering practice, measurements are generally limited due to the limitations of budget and time so that the estimation of the mean value can have uncertainty to some extent. As the Bayesian theory can effectively combine the measurements and prior information to deal with uncertainty, CRF was implemented with the aid of the Bayesian framework in this study. More importantly, this simulation procedure is achieved as an analytical solution to avoid the time-consuming sampling work. The results show that the proposed method can provide a reasonable estimation about the rockhead depth at various locations against measurement data and as a result, the subjectivity in determining prior mean can be minimized. Finally, both the measurement data and selection of hyper-parameters in the proposed method can affect the simulated rockhead profiles, while the influence of the latter is less significant than that of the former.     

Effectiveness of predicting tunneling-induced ground settlements using machine learning methods with small datasets

Prediction of tunneling-induced ground settlements is an essential task, particularly for tunneling in urban settings. Ground settlements should be limited within a tolerable threshold to avoid damages to aboveground structures. Machine learning(ML) methods are becoming popular in many fields, including tunneling and underground excavations, as a powerful learning and predicting technique. However, the available datasets collected from a tunneling project are usually small from the perspective o...     

Adaptive sampling strategy for characterizing spatial distribution of soil liquefaction potential using cone penetration test

Characterizing spatial distribution of soil liquefaction potential is critical for assessing liquefaction-related hazards (e.g. building damages caused by liquefaction-induced differential settlement). However, in engineering practice, soil liquefaction potential is usually measured at limited locations in a specific site using in situ tests, e.g. cone penetration tests (CPTs), due to the restrictions of time, cost and access to subsurface space. In these cases, liquefaction potential of soil at untested locations requires to be interpreted from limited measured data points using proper interpolation method, leading to remarkable statistical uncertainty in liquefaction assessment. This underlines an important question of how to optimize the locations of CPT soundings and determine the minimum number of CPTs for achieving a target reliability level of liquefaction assessment. To tackle this issue, this study proposes a smart sampling strategy for determining the minimum number of CPTs and their optimal locations in a self-adaptive and data-driven manner. The proposed sampling strategy leverages on information entropy and Bayesian compressive sampling (BCS). Both simulated and real CPT data are used to demonstrate the proposed method. Illustrative examples indicate that the proposed method can adaptively and sequentially select the required number and optimal locations of CPTs.     

Evaluating Model Uncertainty of an SPT-based Simplified Method for Reliability Analysis for Probability of Liquefaction

In this paper, an innovative procedure is developed for estimating the uncertainty of an empirical geotechnical model. Here, the Youd et al. (2001) method, a deterministic model for liquefaction triggering evaluation, is examined for its model uncertainty. The procedure for evaluating this model uncertainty involves two steps: 1) deriving a Bayesian mapping function based on a database of case histories, and 2) using the calibrated Bayesian mapping function as a reference to back-figure the uncertainty of the model. Details of the developed procedure within the framework of the first-order reliability method (FORM) are presented. Using FORM with the calibrated model uncertainty, the probability of liquefaction can be readily determined, and thus, the results presented in this paper extend the use of the Youd et al. (2001) method.     

Spatially variable soils affecting geotechnical strip foundation design

Natural soil variability is a well-known issue in geotechnical design, although not frequently managed in practice. When subsoil must be characterized in terms of mechanical properties for infrastructure design, random finite element method (RFEM) can be effectively adopted for shallow foundation design to gain a twofold purpose: (1) understanding how much the bearing capacity is affected by the spatial variability structure of soils, and (2) optimisation of the foundation dimension (i.e. width B). The present study focuses on calculating the bearing capacity of shallow foundations by RFEM in terms of undrained and drained conditions. The spatial variability structure of soil is characterized by the autocorrelation function and the scale of fluctuation (δ). The latter has been derived by geostatistical tools such as the ordinary Kriging (OK) approach based on 182 cone penetration tests (CPTs) performed in the alluvial plain in Bologna Province, Italy. Results show that the increase of the B/δ ratio not only reduces the bearing capacity uncertainty but also increases its mean value under drained conditions. Conversely, under the undrained condition, the autocorrelation function strongly affects the mean values of bearing capacity. Therefore, the authors advise caution when selecting the autocorrelation function model for describing the soil spatial variability structure and point out that undrained conditions are more affected by soil variability compared to the drained ones.     

Machine learning-based automatic control of tunneling posture of shield machine

For a tunnel driven by a shield machine, the posture of the driving machine is essential to the construction quality and environmental impact. However, the machine posture is controlled by the experienced driver of shield machine by setting hundreds of tunneling parameters empirically. Machine learning(ML) algorithm is an alternative method that can let the computer to learn from the driver’s operation and try to model the relationship between parameters automatically. Thus, in this paper, three...     

Bayesian data analysis to quantify the uncertainty of intact rock strength

One of the main difficulties in the geotechnical design process lies in dealing with uncertainty. Uncertainty is associated with natural variation of properties, and the imprecision and unpredictability caused by insufficient information on parameters or models. Probabilistic methods are normally used to quantify uncertainty. However, the frequentist approach commonly used for this purpose has some drawbacks.First, it lacks a formal framework for incorporating knowledge not represented by data. Second, it has limitations in providing a proper measure of the confidence of parameters inferred from data. The Bayesian approach offers a better framework for treating uncertainty in geotechnical design. The advantages of the Bayesian approach for uncertainty quantification are highlighted in this paper with the Bayesian regression analysis of laboratory test data to infer the intact rock strength parameters σ_(ci) and m_i used in the Hoek-Brown strength criterion. Two case examples are used to illustrate different aspects of the Bayesian methodology and to contrast the approach with a frequentist approach represented by the nonlinear least squares(NLLS) method. The paper discusses the use of a Student's t-distribution versus a normal distribution to handle outliers, the consideration of absolute versus relative residuals, and the comparison of quality of fitting results based on standard errors and Bayes factors. Uncertainty quantification with confidence and prediction intervals of the frequentist approach is compared with that based on scatter plots and bands of fitted envelopes of the Bayesian approach. Finally, the Bayesian method is extended to consider two improvements of the fitting analysis. The first is the case in which the Hoek-Brown parameter, a, is treated as a variable to improve the fitting in the triaxial region. The second is the incorporation of the uncertainty in the estimation of the direct tensile strength from Brazilian test results within the overall evaluation of the intact rock strength.     

Comparison of machine learning methods for ground settlement prediction with different tunneling datasets

This study integrates different machine learning (ML) methods and 5-fold cross-validation (CV) method to estimate the ground maximal surface settlement (MSS) induced by tunneling. We further investigate the applicability of artificial intelligent (AI) based prediction through a comparative study of two tunnelling datasets with different sizes and features. Four different ML approaches, including support vector machine (SVM), random forest (RF), back-propagation neural network (BPNN), and deep neural network (DNN), are utilized. Two techniques, i.e. particle swarm optimization (PSO) and grid search (GS) methods, are adopted for hyperparameter optimization. To assess the reliability and efficiency of the predictions, three performance evaluation indicators, including the mean absolute error (MAE), root mean square error (RMSE), and Pearson correlation coefficient (R), are calculated. Our results indicate that proposed models can accurately and efficiently predict the settlement, while the RF model outperforms the other three methods on both datasets. The difference in model performance on two datasets (Datasets A and B) reveals the importance of data quality and quantity. Sensitivity analysis indicates that Dataset A is more significantly affected by geological conditions, while geometric characteristics play a more dominant role on Dataset B.     

Three and more dimensional modelling in geo-engineering

Full three-dimensional modelling has been developed and is implemented for many sites where engineering structures are built. Such computer models of the subsurface allow for a more sophisticated handling of subsurface data leading to, for example, better dimensioning of geotechnical units, the evaluation of hazard and risk, foundation design, tunnel routing, planning and building, etc. Other applications are the back-analysis for completed civil engineering projects to verify the correctness of assumed and estimated ground models and parameters, the verification of the correctness of constitutive models for ground behaviour and the use of back analysis to improve building methodologies or equipment. The paper illustrates some of these advantages with a number of state-of-the-art applications of three-dimensional modelling in engineering geology and geotechnical engineering, highlighting a number of key issues when computer-aided 3D modelling is used: the definition of geotechnical (homogeneous) zones, scale and detail, uncertainty and likelihood of the developed model.     

Back-analysis of geotechnical parameters on PVD-improved ground in the Mekong Delta

This study presents a back-analysis of geotechnical parameters on prefabricated vertical drain improved ground at a site in the Mekong Delta. Various time?settlement behaviors that reflected different clay thicknesses and loading patterns were observed. The total surface settlement behavior at several monitoring locations was simulated using an updated exponential method that considered staged construction. The analyzed results were validated by substituting the values into a theoretical solution for radial consolidation. The estimated theoretical behaviors were comparable with the monitored behaviors. The geotechnical parameters were back-analyzed by applying the previously analyzed results to various theoretical and empirical formulas. However, the use of extensometer data that were installed at large intervals produced different values of the geotechnical properties. Furthermore, finite element analysis supported the back-analyzed total settlement behaviors and nearly disregarded the application of the geotechnical properties that were obtained using either surface or subsurface settlement data. However, settlements and excess pore pressures in the sublayers were not successfully predicted even when the geotechnical properties were adjusted. Thus, subsurface instruments that can be installed closely in thick clay deposits are required to reliably reevaluate the variations in geotechnical properties along a certain depth.     

玉溪红塔地块地基处理与效果检验

为了解某地基稳定性及地基的承载能力，对场地地基土层、地质岩性等结合不同作业点之间的联系与区别作了深入对比分析，通过方案比选，选择强夯法作为其最终的地基处理方法，采用规范填料对地基夯实加固有明显的效果，建议填料应严格按照设计要求选用且合理地增加触探作业点。     

Project geotechnical engineering: systems methodology considerations

Systems methodology considerations for project geotechnical engineering are discussed. Emphasis is on general systems and related geotechnical principles, issues and limitations. After an introduction, the paper outlines geotechnical programme management as a framework for defining, organizing and controlling the geotechnical engineering process. Then, following the logical order of elements in project geotechnical engineering, the paper discusses programme development, followed by site characterization and analysis, and design and implementation. The paper concludes with a discussion on additional general methodologies: uncertainty, updating, values, innovation and impediments. The objective is to provide a broad conceptual framework which can be used as needed to help guide realistic integration and implementation of the diverse and often convoluted efforts of project geotechnical engineering.     

Combination of Kriging methods and multi-fractal analysis for estimating spatial distribution of geotechnical parameters

Spatial variability (randomness, correlation, and singularity) within the geotechnical parameters of complicated geological movements influences the estimation quality that depends on how well mathematical tools can account for variability through limited observations of a spatial field. Classical statistical methods depict randomness well, but cannot account for the problems associated with spatial correlations. Geostatistical methods such as ordinary Kriging (OK), universal Kriging (UK), and co-Kriging (CK) can produce predictions based on spatial auto-correlation and cross-correlation, but are always accompanied by average smoothing effects; a local singularity created by nonlinear geo-processes, therefore, requires special methods to be properly evaluated. In this study, a shallow load-bearing stratum of silt clay (length = 525 m, width = 80 m) at the former 2010 Expo Park in Shanghai was explored by performing 42 borehole laboratory experiments, which provided the key geotechnical parameters: the cohesion coefficient (\( C \), in kPa), the friction angle (\( \varphi \), in ^o), and the compression modulus (\( E_{\text{S}} \), in MPa). First, Kriging methods such as OK, UK, and CK estimated these geotechnical parameters, then a multi-fractal analysis was employed to measure the local singularity. Cross-validation illustrates that multi-fractal analysis has the ability to depict a local anomaly, and further that the auxiliary information utilized in CK improves spatial estimation accuracy.     

Fuzzy Models in Geotechnical Engineering and Construction Management

This article is devoted to a variety of applications of fuzzy models in civil engineering, presenting the current work of a group of researchers at the University of Innsbruck. With fuzzy methods and possibility theory as an encompassing framework, the following areas are addressed: uncertainties in geotechnical engineering, fuzzy finite‐element computation of a foundation raft, fuzzy dynamic systems, and processing uncertainty in project scheduling and cost planning.     

Data-driven estimation of joint roughness coefficient

Joint roughness is one of the most important issues in the hydromechanical behavior of rock mass. Therefore, the joint roughness coefficient (JRC) estimation is of paramount importance in geomechanics engineering applications. Studies show that the application of statistical parameters alone may not produce a sufficiently reliable estimation of the JRC values. Therefore, alternative data-driven methods are proposed to assess the JRC values. In this study, Gaussian process (GP), K-star, random forest (RF), and extreme gradient boosting (XGBoost) models are employed, and their performance and accuracy are compared with those of benchmark regression formula (i.e. Z₂, R_p, and SD_i) for the JRC estimation. To analyze the models’ performance, 112 rock joint profile datasets having eight common statistical parameters (R_ave, R_max, SD_h, i_ave, SD_i, Z₂, R_p, and SF) and one output variable (JRC) are utilized, of which 89 and 23 datasets are used for training and validation of models, respectively. The interpretability of the developed XGBoost model is presented in terms of feature importance ranking, partial dependence plots (PDPs), feature interaction, and local interpretable model-agnostic explanations (LIME) techniques. Analyses of results show that machine learning models demonstrate higher accuracy and precision for estimating JRC values compared with the benchmark empirical equations, indicating the generalization ability of the data-driven models in better estimation accuracy.     

Structural performance assessment of GFRP elastic gridshells by machine learning interpretability methods

The prediction of structural performance plays a significant role in damage assessment of glass fiber reinforcement polymer (GFRP) elastic gridshell structures. Machine learning (ML) approaches are implemented in this study, to predict maximum stress and displacement of GFRP elastic gridshell structures. Several ML algorithms, including linear regression (LR), ridge regression (RR), support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), category boosting (CatBoost), and light gradient boosting machine (LightGBM), are implemented in this study. Output features of structural performance considered in this study are the maximum stress as f₁(x) and the maximum displacement to self-weight ratio as f₂(x). A comparative study is conducted and the Catboost model presents the highest prediction accuracy. Finally, interpretable ML approaches, including shapely additive explanations (SHAP), partial dependence plot (PDP), and accumulated local effects (ALE), are applied to explain the predictions. SHAP is employed to describe the importance of each variable to structural performance both locally and globally. The results of sensitivity analysis (SA), feature importance of the CatBoost model and SHAP approach indicate the same parameters as the most significant variables for f₁(x) and f₂(x).     

Stratigraphic uncertainty characterization using generalized coupled Markov chain

Stratigraphic uncertainty widely exists in geotechnical structures. An efficient generalized coupled Markov chain (GCMC) model was proposed to simulate the stratigraphic uncertainty in the past. This model, however, cannot be directly applied to geotechnical problems because of the difficulty in directly estimating the horizontal transition probability matrices (HTPMs) of the GCMC from boreholes with large hole-to-hole distance. This paper, therefore, aims at addressing the above problem by using a maximum likelihood estimation (MLE) method and utilizing the MLE modified GCMC model to characterize stratigraphic uncertainty. In the framework of the GCMC, the validity of the proposed method is verified, and information entropy method is adopted to quantify the stratigraphic uncertainty. As an illustration, the proposed model is applied to a construction site in Hong Kong to characterize the stratigraphic uncertainty. A systematic comparison of coupled Markov chain (CMC) and GCMC model for stratigraphic uncertainty simulation was conducted. The influences of borehole layout scheme on transition probability matrices (TPMs) estimation and stratigraphic uncertainty simulation are explored. Results showed that the result of stratigraphic uncertainty simulation of the GCMC model is more reasonable than that of the CMC model. The borehole layout scheme is very important for estimating TPMs and significantly influences the stratigraphic uncertainty simulation. The information entropy map has the potential to become a useful guideline for the design of the borehole layout scheme in geotechnical engineering practice.