Prediction Of Pile Bearing Capacity Research Articles

Optimizing pile bearing capacity prediction: Insights from dynamic testing and smart algorithms in geotechnical engineering

In contemporary mining and geotechnical projects, various approaches are employed to predict the bearing capacity of piles (Qu). However, accurately modeling pile behavior using numerical, experimental, analytical, and regression methods proves challenging and, at times, infeasible due to the intricate nature of geotechnical materials, uncertainties, and the interaction between soil and piles. Consequently, formulations are generally presented, incorporating assumptions and simplifications that deviate from the actual complexity of the problem. Moreover, despite the high accuracy of pile loading tests as a reliable method in various design stages, their high costs and time requirements deter designers from conducting field tests. To address these challenges, this study performed 50 dynamic tests (HSDT) on precast concrete piles in Indonesia, Pekanbaru, to generate the necessary datasets. To mitigate experimental costs, two optimization algorithms fruit fly optimization (FFO) and invasive weed optimization (IWO) were employed. These models incorporated pile set (S), drop height (H), hammer weight (W), cross-sectional area (A), and length (L) as input parameters. Finally, the models' accuracy was evaluated using squared correlation coefficient (R2), variance accounted for (VAF), mean square error (MSE), mean absolute percentage error (MAPE), and root mean square error (RMSE). The results revealed that the FFO algorithm achieved an accuracy range of 0.971–0.978. Similarly, the IWO algorithm exhibited an accuracy range of 0.984–0.988. Additionally, sensitivity analysis indicated that, among the input parameters, W had the most significant impact on the Qu in both algorithms.

Prediction of pile-bearing capacity using Least Square Support Vector Regression: individual and hybrid models development

Prediction of pile-bearing capacity using Least Square Support Vector Regression: individual and hybrid models development

Probabilistic analysis of seasonal influence on the prediction of pile bearing capacity by CPT: A case study

In this study, we used a probabilistic approach to evaluate the probability of failure of small diameter bored piles in unsaturated sandy soils. The prediction of the bearing capacity was based on the Cone Penetration Test (CPT). We assessed the seasonal influence by analyzing CPT data from dry and wet seasons. The LCPC and Aoki-Velloso semiempirical methods were deterministically applied, and the Aoki-Velloso method was examined using reliability theory. The first-order reliability method (FORM) was used to evaluate the probability of failure and calculate the reliability index based on random variables obtained from CPT data considering the effect of season when predicting bearing capacity. A total of 8.0 m long bored piles with diameters of 0.25, 0.30, and 0.35 m were used for both deterministic and probabilistic analyses. We also evaluated the influence of load ratio variation on the reliability index and determined the pile length required to reach a target reliability index value. Reliability analyses considering CPT campaigns conducted in different seasons showed the effect of season on the prediction of bearing capacity. For example, an 8.0 m long pile with a 0.30 m diameter and 100 kN vertical load had a probability of failure of 38.7 % in the wet season and 0.01 % in the dry season. The reliability analyses provided insights into the influence of seasonal variability on in situ tests for pile bearing capacity, affecting the design of foundation elements.

Alternative Method of Calculation of Pile Bearing Capacity Using Graphical Correction N-SPT

Field N-SPT, used as the primary data in calculating pile bearing capacity, must be corrected by analytical methods to become N-SPT analytical correction. The analytical process requires unit weight, groundwater level, soil classification, and drilling equipment specifications to calculate field N-SPT to become analytical correction of N-SPT. A graphical approach is proposed to simplify the analytical method and to accelerate calculating the estimated pile-bearing capacity, which can be directly obtained from the relationship between the field N-SPT and the graphical correction N-SPT. The calculation results of analytically corrected N-SPT and graphical corrected N-SPT based on the field N-SPT show that the graphic of N-SPT increases linearly acording value of the field N-SPT. From all the calculations of the corrected N-SPT and pile bearing capacity obtained the ratio of pile capacity (RQ) of the graphical corrected N-SPT to the analytically corrected N-SPT are RQu = 0.93, RQp = 0.94, and RQs = 0,94. This shows that the N-SPT Graphical Correction chart can be used to simplify and accelerate the prediction of pile-bearing capacity based on field N-SPT

Prediction and Optimization of Pile Bearing Capacity Considering Effects of Time

Prediction of pile bearing capacity has been considered an unsolved problem for years. This study presents a practical solution for the preparation and maximization of pile bearing capacity, considering the effects of time after the end of pile driving. The prediction phase proposes an intelligent equation using a genetic programming (GP) model. Thus, pile geometry, soil properties, initial pile capacity, and time after the end of driving were considered predictors to predict pile bearing capacity. The developed GP equation provided an acceptable level of accuracy in estimating pile bearing capacity. In the optimization phase, the developed GP equation was used as input in two powerful optimization algorithms, namely, the artificial bee colony (ABC) and the grey wolf optimization (GWO), in order to obtain the highest bearing capacity of the pile, which corresponds to the optimum values for input parameters. Among these two algorithms, GWO obtained a higher value for pile capacity compared to the ABC algorithm. The introduced models and their modeling procedure in this study can be used to predict the ultimate capacity of piles in such projects.

Prediction of Pile Bearing Capacity Using Opposition-Based Differential Flower Pollination-Optimized Least Squares Support Vector Regression (ODFP-LSSVR)

Pile foundations are widely used for high-rise structures constructed in soft ground. The bearing capacity of pile is a crucial parameter required during the design and construction phase of pile foundation engineering projects. In practice, accurate predictions of pile bearing capacity are challenging due to a complex interplay of various geotechnical engineering factors including pile characteristics and ground conditions. This study proposes a data-driven model for coping with the problem of interest that hybridizes machine learning and metaheuristic approaches. Least squares support vector regression (LSSVR) is used for analyzing a dataset containing historical records of pile tests. Based on such datasets, LSSVR is capable of generalizing a multivariate function that estimates values of pile bearing capacity based on a set of variables describing pile characteristics and ground conditions. Moreover, opposition-based differential flower pollination (ODFP) metaheuristic is proposed to optimize the LSSVR learning process. Experimental results supported by the statistical test showed that the proposed ODFP-optimized LSSVR can achieve a good predictive performance in terms of root mean square error, mean absolute percentage error mean absolute error, and coefficient of determination. These results confirm that the ODFP-optimized LSSVR can be a potential alternative to assist civil engineers in the task of pile bearing capacity estimation.

Prediction of Pile Bearing Capacity Using XGBoost Algorithm: Modeling and Performance Evaluation

The major criteria that control pile foundation design is pile bearing capacity (Pu). The load bearing capacity of piles is affected by the various characteristics of soils and the involvement of multiple parameters related to both soil and foundation. In this study, a new model for predicting bearing capacity is developed using an extreme gradient boosting (XGBoost) algorithm. A total of 200 driven piles static load test-based case histories were used to construct and verify the model. The developed XGBoost model results were compared to a number of commonly used algorithms—Adaptive Boosting (AdaBoost), Random Forest (RF), Decision Tree (DT) and Support Vector Machine (SVM) using various performance measure metrics such as coefficient of determination, mean absolute error, root mean square error, mean absolute relative error, Nash–Sutcliffe model efficiency coefficient and relative strength ratio. Furthermore, sensitivity analysis was performed to determine the effect of input parameters on Pu. The results show that all of the developed models were capable of making accurate predictions however the XGBoost algorithm surpasses others, followed by AdaBoost, RF, DT, and SVM. The sensitivity analysis result shows that the SPT blow count along the pile shaft has the greatest effect on the Pu.

Application of Ensemble Learning Using Weight Voting Protocol in the Prediction of Pile Bearing Capacity

Accurate prediction of pile bearing capacity is an important part of foundation engineering. Notably, the determination of pile bearing capacity through an in situ load test is costly and time-consuming. Therefore, this study focused on developing a machine learning algorithm, namely, Ensemble Learning (EL), using weight voting protocol of three base machine learning algorithms, gradient boosting (GB), random forest (RF), and classic linear regression (LR), to predict the bearing capacity of the pile. Data includes 108 pile load tests under different conditions used for model training and testing. Performance evaluation indicators such as R-square (R2), root mean square error (RMSE), and MAE (mean absolute error) were used to evaluate the performance of models showing the efficiency of predicting pile bearing capacity with outstanding performance compared to other models. The results also showed that the EL model with a weight combination of w 1 = 0.482, w 2 = 0.338, and w 3 = 0.18 corresponding to the models GB, RF, and LR gave the best performance and achieved the best balance on all data sets. In addition, the global sensitivity analysis technique was used to detect the most important input features in determining the bearing capacity of the pile. This study provides an effective tool to predict pile load capacity with expert performance.

Application of Developed New Artificial Intelligence Approaches in Civil Engineering for Ultimate Pile Bearing Capacity Prediction in Soil Based on Experimental Datasets

In this study, a neural-fuzzy (NF) system is combined with group method of data handling (GMDH) in order to estimate the axial bearing capacity of driven piles. To reach optimum design of this conjunction (NF-GMDH) network, the metaheuristic techniques including particle swarm optimization (PSO) and gravitational search algorithm (GSA) were utilized. The datasets used for estimating pile bearing capacity were collected from the literature review. The parameters influencing the modeling and pile capacity analysis were taken into account as Flap number, surrounding soil properties, the pile geometric characteristics, and internal friction angles of the pile–soil interface. The efficiency of hybrid NF-GMDH networks in train and test phases was examined. Applying the PSO algorithm to the hybrid NF-GMDH model structure improved the model performance and achieved a higher level of accuracy in predicting the ultimate pile bearing capacity (RMSE = 1375 and SI = 0.255) compared to NF-GMDH model developed by GSA (RMSE = 1740.7 and SI = 0.357). In addition, based on achieved results, the developed NF-GMDH networks showed relatively better performances in comparison with gene programming and linear regression model methods considered in this study.

Neuro-genetic, neuro-imperialism and genetic programing models in predicting ultimate bearing capacity of pile

Prediction of pile-bearing capacity developing artificial intelligence models has been done over the last decade. Such predictive tools can assist geotechnical engineers to easily determine the ultimate pile bearing capacity instead of conducting any difficult field tests. The main aim of this study is to predict the bearing capacity of pile developing several smart models, i.e., neuro-genetic, neuro-imperialism, genetic programing (GP) and artificial neural network (ANN). For this purpose, a number of concrete pile characteristics and its dynamic load test specifications were investigated to select pile cross-sectional area, pile length, pile set, hammer weight and drop height as five input variables which have the most impacts on pile bearing capacity as the single output variable. It should be noted that all the aforementioned parameters were measured by conducting a series of pile driving analyzer tests on precast concrete piles located in Pekanbaru, Indonesia. The recorded data were used to establish a database of 50 test cases. With regard to data modelling, many smart models of neuro-genetic, neuro-imperialism, GP and ANN were developed and then evaluated based on the three most common statistical indices, i.e., root mean squared error (RMSE), coefficient determination (R2) and variance account for (VAF). Based on the simulation results and the computed indices’ values, it is observed that the proposed GP model with training and test RMSE values of 0.041 and 0.040, respectively, performs noticeably better than the proposed neuro-genetic model with RMSE values of 0.042 and 0.040, neuro-imperialism model with RMSE values of 0.045 and 0.059, and ANN model with RMSE values of 0.116 and 0.108 for training and test sets, respectively. Therefore, this GP-based model can provide a new applicable equation to effectively predict the ultimate pile bearing capacity.

Differential Settlement of Intersecting Buildings in an Offshore Reclamation Project

This paper addresses the problem of open gaps caused by differential settlement in the process of constructing sluice buildings in soft soil beach areas, combined with the construction of sluice and supporting facilities in a reclamation project. First, the change rules for the shear strength and compression modulus of soft soil under different consolidation degrees are studied by theoretical analysis. Then, an interaction model for soft soil and pile soil is established using the geotechnical finite element analysis software MIDAS/GTS NX. The change rules for the vertical and horizontal ultimate bearing capacities of a single pile with the degree of soil consolidation are studied. On this basis, a three-dimensional numerical analysis model of drainage sluice, seawall, cofferdam, and foundation soil is established, and the relationship between the degree of soil consolidation and the development of structural gaps caused by differential settlement is obtained. The research results show that the bearing capacity of a single pile increases greatly with the consolidation of soil around the pile and that the gap width between the structures in the project decreases with increasing consolidation. This paper provides a theoretical basis for the prediction of pile bearing capacity in the preliminary design stage and the evaluation and calculation of differential settlement of intersecting buildings in soft soil beach areas.

The effects of particle swarm optimization and genetic algorithm on ANN results in predicting pile bearing capacity

Pile as a foundation type has transferred the heavy structural loads into the ground. Proper prediction and determining of pile bearing capacity has been signified in initial geotechnical structures designing. The current study has attempted to build two hybrid intelligent models for pile bearing capacity prediction. Presenting the influence of genetic algorithm (GA) and particle swarm optimisation (PSO) on a pre-developed artificial neural network (ANN), two hybrid models, i.e., GA-ANN and PSO-ANN have been built to pile bearing capacity prediction. To do this, an established database in literature has been used to develop intelligent systems. Then, the most important parameters on GA and PSO have been designed to optimise ANN weights and biases for receiving the best results. Then, the best predictive models of GA-ANN and PSO-ANN were selected based on three performance indices, i.e., R2, RMSE and VAF. Respectively, R2 variables as (0.975 and 0.988) and (0.985 and 0.993) have been gained to train and test of datasets in GA-ANN and PSO-ANN. The outcomes have proved both hybrid methods as capable with highly accurate bearing capacity prediction, however, PSO-ANN predictive model is more applicable in terms of performance capacity and it can be introduced as a new technique in this field.

Applying several soft computing techniques for prediction of bearing capacity of driven piles

Pile as a type of foundation is a structure which can transfer heavy structural loads into the ground. Determination and proper prediction of pile bearing capacity are considered as a very important issue in preliminary design of geotechnical structures. This study attempts to develop several intelligent techniques for prediction of pile bearing capacity in cohesionless soil. To show the effects of fuzzy inference system and imperialism competitive algorithm (ICA) on a pre-developed artificial neural network (ANN), two hybrid ANN models namely ICA-ANN and adoptive neuro-fuzzy inference system (ANFIS) were considered and developed to estimate pile bearing capacity. Then, results of these techniques were compared with those of ANN model and the best one among them was chosen according to the results of performance indices. Several parameters (i.e., internal friction angle of soil located in shaft and tip, effective vertical stress at pile toe, pile area, and pile length) were set as model inputs, while the output is the total driven pile bearing capacity. As a result of the developed models, coefficient of determination (R2) values of (0.895, 0.905), (0.945, 0.958), and (0.967, 0.975) were obtained for training and testing data sets of ANN, ICA-ANN, and ANFIS models, respectively. The results showed that both hybrid models are able to predict bearing capacity with high degree of accuracy; however, ANFIS receives more applicable based on used performance indices and it can be utilized for further researchers and engineers in practice.

Regressive approach for predicting bearing capacity of bored piles from cone penetration test data

In this study, the least square support vector machine (LSSVM) algorithm was applied to predicting the bearing capacity of bored piles embedded in sand and mixed soils. Pile geometry and cone penetration test (CPT) results were used as input variables for prediction of pile bearing capacity. The data used were collected from the existing literature and consisted of 50 case records. The application of LSSVM was carried out by dividing the data into three sets: a training set for learning the problem and obtaining a relationship between input variables and pile bearing capacity, and testing and validation sets for evaluation of the predictive and generalization ability of the obtained relationship. The predictions of pile bearing capacity by LSSVM were evaluated by comparing with experimental data and with those by traditional CPT-based methods and the gene expression programming (GEP) model. It was found that the LSSVM performs well with coefficient of determination, mean, and standard deviation equivalent to 0.99, 1.03, and 0.08, respectively, for the testing set, and 1, 1.04, and 0.11, respectively, for the validation set. The low values of the calculated mean squared error and mean absolute error indicated that the LSSVM was accurate in predicting the pile bearing capacity. The results of comparison also showed that the proposed algorithm predicted the pile bearing capacity more accurately than the traditional methods including the GEP model.

Reliability assessment on prediction of pile bearing capacity

There are many prediction methods that can be used to determine the bearing capacity of pipe pile. However, most of these methods are highly sensitive to specific in situ conditions and hence usually result in significant error. Therefore, it is important to evaluate the reliability of different pile bearing capacity prediction methods. Based on reliability theory, a computer code is developed in this paper to calculate reliability. Random variable statistic properties and corresponding reliabilities for different prediction methods are investigated and analysed. Results show that the reliability index for logarithmic normal distribution is higher than normal distribution. Analysis results demonstrate that the checking point method (JC method) method has a slightly lower efficiency than the Monte-Carlo importance sampling method and the Monte-Carlo direct sampling method. The Monte-Carlo importance sampling method has a higher efficiency than the Monte-Carlo direct sampling method. The Monte-Carlo importance sampling method can be used in major engineering projects. It is concluded that the clustering centre method has higher reliability in predicting pile bearing capacity than the Eslami and Fellenius method, Almeida method and Powell method.

Prediction of pile bearing capacity using a hybrid genetic algorithm-based ANN

The application of artificial neural network (ANN) in predicting pile bearing capacity is underlined in several studies. However, ANN deficiencies in finding global minima as well as its slow rate of convergence are the major drawbacks of implementing this technique. The current study aimed at developing an ANN-based predictive model enhanced with genetic algorithm (GA) optimization technique to predict the bearing capacity of piles. To provide necessary dataset required for establishing the model, 50 dynamic load tests were conducted on precast concrete piles in Pekanbaru, Indonesia. The pile geometrical properties, pile set, hammer weight and drop height were set to be the network inputs and the pile ultimate bearing capacity was set to be the output of the GA-based ANN model. The best predictive model was selected after conducting a sensitivity analysis for determining the optimum GA parameters coupled with a trial-and-error method for finding the optimum network architecture i.e. number of hidden nodes. Results indicate that the pile bearing capacities predicted by GA-based ANN are in close agreement with measured bearing capacities. Coefficient of determination as well as mean square error equal to 0.990 and 0.002 for testing datasets respectively, suggest that implementation of GA-based ANN models as a highly-reliable, efficient and practical tool in predicting the pile bearing capacity is of advantage.

Effect of spatial correlation of standard penetration test (SPT) data on bearing capacity of driven piles in sand

In this paper, the effect of spatial correlation of standard penetration test (SPT) data on the bearing capacity of driven piles in sand is analyzed. First, the direct approach for using SPT data to determine the bearing capacity of piles in sand is used to derive the expressions for probabilistic prediction of pile bearing capacity by considering the spatial correlation of the SPT data. To analyze the relationship between the probability of failure and the factor of safety, a procedure based on the advanced first-order, second-moment (FOSM) method is used. Then parametric studies are conducted on the spatial correlation between the spatial average of SPT numbers over the pile length, NLV, and the spatial average of SPT numbers over an interval near the pile base, NbV, and its effect on the bearing capacity of piles. The results indicate that it is important to consider the spatial correlation between NLVand NbVin the probabilistic prediction of pile bearing capacity. Ignoring this spatial correlation will underestimate the probability of failure and lead to unsafe design. Finally, three tested piles are analyzed to demonstrate the probabilistic analysis of piles by considering the spatial correlation of SPT data and the procedure for probabilistic analysis of pile bearing capacity is summarized.

Prediction of pile bearing capacity using support vector machine

This paper examines the potential of support vector machine (SVM) in prediction of bearing capacity of pile from pile load test data. In this study, SVM uses regression technique by introducing ε-insensitive loss function. The data from a pile load test has been used to build the SVM model. SVM uses penetration depth ratio (<i>l/d</i>), mean normal stress (σ_<i>m</i>), and no of blows (<i>n</i>) as input parameters. The output of SVM model is bearing capacity (<i>Q</i>) of pile. Sensitivity analysis of the develop SVM model shows that <i>l/d</i> has the most significant effect on <i>Q</i>. An equation has been also developed for the prediction of bearing capacity of pile. This study shows that SVM approach give an alternative tools to geotechnical engineers for the determination of bearing capacity of pile.

Prediction of pile bearing capacity using artificial neural networks

It is well known that the human brain has the advantage of handling disperse and parallel distributed data efficiently. On the basis of this fact, artificial neural networks theory was developed and has been applied to various fields of science successfully. In this study, error back propagation neural networks were utilized to predict the ultimate bearing capacity of piles. For the verification of applicability of neural networks, results of model pile load tests performed by the authors were simulated. In addition, the results of in situ pile load tests obtained from a literature survey were also used. The results showed that the maximum error of prediction did not exceed 25%, except for some bias data. These limited results indicated the feasibility of utilizing neural networks for pile capacity prediction problems.

Prediction of pile capacity from stress‐wave measurements: A neural network approach

AbstractA neural network approach for the prediction of pile bearing capacity by the stress‐wave matching technique is presented. The main advantage of this approach over the traditional manual or automated matching approach is that it avoids the time‐consuming process of iterative adjustment. This makes it feasible to determine the static pile capacity in real time in the field. Another benefit of this approach is that as more case histories become available, the neural network can be improved by learning from these new examples.A three‐layer back‐propagation network is set up to illustrate the capability of the proposed approach for 70 dynamically tested concrete bored piles. A wave equation model developed at the National University of Singapore and coded in the NUSWAP computer program is used to formulate the problem. Up to 14 of the 70 piles (20 percent) are used in training the network. The NUSWAP program is used to generate simulation training examples based on the manually fitted training examples for further training of the network. Different network configurations are examined. The trained network produces results exhibiting good stress‐wave matching qualities compared to those obtained by manual fitting. The pile bearing capacities predicted by the two approaches agree very closely. The load‐settlement curve and axial load distribution in the pile computed using the network‐predicted soil parameters are in good agreement with the field measurements obtained from a maintained load test.