Explore the vast range of titles available.

Determination of pile bearing capacity is essential in pile foundation design. This study focused on the use of evolutionary algorithms to optimize Deep Learning Neural Network (DLNN) algorithm to predict the bearing capacity of driven pile. For this purpose, a Genetic Algorithm (GA) was developed to select the most significant features in the raw dataset. After that, a GA-DLNN hybrid model was developed to select optimal parameters for the DLNN model, including: network algorithm, activation function for hidden neurons, number of hidden layers, and the number of neurons in each hidden layer. A database containing 472 driven pile static load test reports was used. The dataset was divided into three parts, namely the training set (60%), validation (20%) and testing set (20%) for the construction, validation and testing phases of the proposed model, respectively. Various quality assessment criteria, namely the coefficient of determination (R 2 ), Index of Agreement (IA), mean absolute error (MAE) and root mean squared error (RMSE), were used to evaluate the performance of the machine learning (ML) algorithms. The GA-DLNN hybrid model was shown to exhibit the ability to find the most optimal set of parameters for the prediction process.The results showed that the performance of the hybrid model using only the most critical features gave the highest accuracy, compared with those obtained by the hybrid model using all input variables.

This paper aims to investigate the effect of combined end-and-side grouting on the bearing properties of large-diameter rock-socketed bored piles in highly weathered rock layers. Eight full-scale pile load tests were conducted in the highly weathered rock layer to analyze the enhanced mechanism of the combined grouted bored piles. The test data from pile mechanical testing were compared with the recommended values in the current specification and geological survey report. The results demonstrate significant improvement in the side and end resistances of the combined grouted bored piles, resulting in a substantial increase in the bearing capacity and effective settlement control. It was observed that the construction of impact holes for bored piles can cause severe damage to highly weathered rock structures and weaken the mobilization of side and end resistances. Moreover, it was found that the calculation of the enhancement coefficient in the current specification underestimates the practical bearing capacity. The measured enhancement coefficients for the side and end resistance of piles in fully or highly weathered rock layers range from 2.49 to 3.05 and 2.24 to 2.43, respectively, which are more reasonable and feasible for the calculation. The research findings deepen the understanding of the bearing characteristics of large-diameter rock-socketed bored piles with combined grouting and provide valuable case references for the optimal design of large-diameter combined grouted piles for building foundations in Shenzhen, China.HighlightsPost-grouting had the potential to improve super high-building foundation reliability while reducing pile length and cost.The improvement effect and improvement mechanism of combined grouted bored piles embedded in highly rock strata were revealed.The influence of the size effect for large-diameter piles in highly weathered rock was revealed.The construction of impact holes for bored piles can cause severe damage to highly weathered rock structures and weaken the mobilization of side and end resistances.The enhancement coefficients for the side and end resistance of piles in fully or highly weathered rock layers were proposed.

In the process of designing pile foundations, it is essential to take the axial bearing capacity ( B c ) of the pile into consideration., where determination of this target requires extreme fields and experimental efforts along with its cost. The primary objective of this study was to investigate the possibility of using tree-based approaches in order to estimate the axial B c of piles. The goal of building Random Forests ( RF ) models is to produce a strong and adaptable machine learning method that is capable of reliably and accurately completing tasks related to classification as well as regression. Enhanced precision in predicting feature significance, scalability, adaptability, and managing missing data are the main objectives of employing RF . The accuracy of this model is very dependent on its hyperparameters, which are linked to the Coati optimizer ( CO ) and Giant trevally optimizer ( GTO ) procedures (also called R F - C and R F - G ) in order to find the optimal combinations. In a database, there were 472 driven pile static load test results collected from previous papers. Specifically, the construction, validation, and testing phases of the proposed framework were carried out using the learning set (70%), validation set (15%), and evaluating set (15%) of the dataset. Moreover, the feature importance analysis is designed to assess the impact of each input variable on the axial B c of piles. R F - C and R F - G offer promising B c forecasting capabilities, where the R F - C approach outperformed the R F - G method in R 2 values, with values of 0.9876, 0.9781, and 0.9873.

Rock failure is a serious problem in rock engineering environments. Rock failure modes, however, are complex and difficult to quantify or predict. A comprehensive study on rock failure modes at laboratory scale is, therefore, potentially important as it helps recognize the adequacy of the support designed on the basis of the nature of an engineering work. With due need, this paper analyzes the failure modes of granite, schist, and sandstone under uniaxial compression, Brazilian, and point load tests in relation to corresponding strengths. The nature of the principal failure mode changes from axial splitting to shearing along a single plane to multiple fracturing in the case of both granite and sandstone specimens as uniaxial compressive strength (UCS) increases. In the case of schist, specimens failed at low UCS show failure along foliations whereas specimens which do not fail along foliations portray high strength. The relation between failure modes of all three rocks under uniaxial compression and corresponding UCS values was broadly explained in terms of damage evolution of the rocks. Granite and sandstone specimens failed mainly following central or central multiple type of fracturing whereas schist specimens principally failed by layer activation in combination with either central or non-central fractures over the entire range of determined Brazilian tensile strength. In the case of granite and sandstone, central multiple failure mode corresponds to high tensile strength. Descriptions of different failure modes under point loading were presented. It was found that granite and sandstone specimens generally fail through the rock materials in one or more extensional planes containing the line of loading. Failure patterns showing triple junctions correspond to high point load strength indices. In the case of schist, specimens failed along foliations show a low point load strength index whereas specimens failed through material with a single extensional plane result in high strength.

in geotechnical engineering, foundation piles are ideal for deep foundations that cannot bear higher loads. This architectural expansion places a great deal of responsibility on the engineer to anticipate the appropriate load for the constructor. Unfortunately, calculations of the pile’s bearing capacity are not accessible. It has always been a source of concern for geotechnical engineers, as the structure’s safety depends on the pile’s bearing capacity and gives it a safe value. These research tests are previously known pile load test data from several locations in Nasiriyah to determine the ultimate load-carrying capacity using various interpreting methodologies. A database that was used to test the pile load for three different areas in Nasiriyah, southern Iraq: The Main Drain River Bridge Project, the Al-Eskan Interchange Project, and the Al-Hawra Hospital, as determined by analytical methods, as well as evaluating the final loading values resulting from the methods used, by ASTM D-1143, American and British Standard Code of Practice BS 800. The final capacity for the pile bearing is estimated using these approaches, which are depicted in the form of a graph-based on field data. Chin-Kondner and Brinch Hansen algorithms anticipate the highest failure load for all piles based on the comparison. On average, Chin–Kondner’s ultimate load is 22% higher than Hansen’s maximum load for the 22 pile load tests. Decourt and DeBeer, and Mazurkiewicz’s techniques yielded the closest average failure load. Buttler-Hoy approach yielded the smallest failure load.

Safety has been always challenging in geotechnical engineering owing to the inherently variable nature of the soil. In pile foundations, conducting field tests is highly expensive and time-consuming, and thus soft-computing based simulation models analysis is a realistic and useful alternative. This study presented a comparative analysis of artificial neural network (ANN)-based hybrid models and conventional soft computing techniques to estimate the probability of failure of pile foundation. With this respect, dynamic pile load test data of pile foundations were used to construct ANN-based models. Five widely used meta-heuristic optimization algorithms, namely particle swarm optimization, grasshopper optimization algorithm, artificial bee colony, ant colony optimization, and ant lion optimizer, were employed for this purpose. In addition, three widely used conventional soft computing techniques; including genetic programming (GP), multivariate adaptive regression splines (MARS), and group method of data handling (GMDH) were utilized for comparison purposes. The performances of all the developed models were assessed using various statistical performance indices. Experimental results show that the ANN-PSO (hybrid model of ANN and particle swarm optimization) and GP estimate the probability of failure of pile foundation accurately both in training and testing phases. However, a detailed review of results reveals that the ANN-PSO ( R 2  = 0.9773, RMSE = 0.0439) and GP ( R 2  = 0.9859, RMSE = 0.0353) showed comparatively better performance in the testing phase. The result of the ANN-PSO and GP models is significantly better than those obtained from other benchmark methods. Based on the results, the developed ANN-PSO and GP models can be used to estimate the probability of failure of pile foundation in the design phase of civil engineering projects.

Most currently available load-transfer models for axially loaded pile are either highly empirical or oversimplified without rigorous theoretical manner. This paper presents a novel and rigorous load-transfer model for axially loaded piles installed in modified Cam-clay (MCC) soils under undrained loading. The model is formulated in terms of the Lagrangian description based on the rigorous deformation mechanism of the soil around the pile shaft during undrained loading. Considering the boundary conditions and the equilibrium state of the soil, a t–z curve is extracted from the well-established MCC model that is employed to represent the elastoplastic behaviour of soil surrounding the pile. The proposed model is compared with a well-established finite element model and applied to predict the load–displacement behaviour of two well-documented piles to manifest the validity and capability of the model. The results demonstrate that the present theoretical model can well predict the elastoplastic load–displacement response of the pile and is capable of reflecting some important phenomena observed from pile load tests. The present model provides a rigorous, practical, and effective approach for estimation of the load–displacement behaviour of frictional piles installed in clayey soils.

This paper presents an analysis of long, large-diameter bored piles’ behavior under static and dynamic load tests for a megaproject located in El Alamein, on the northern shoreline of Egypt. Site investigations depict an abundance of limestone fragments and weak argillaceous limestone interlaid with gravelly, silty sands and silty, gravelly clay layers. These layers are classified as intermediate geomaterials, IGMs, and soil layers. The project consists of high-rise buildings founded on long bored piles of 1200 mm and 800 mm in diameter. Forty-four (44) static and dynamic compression load tests were performed in this study. During the pile testing, it was recognized that the pile load–settlement behavior is very conservative. Settlement did not exceed 1.6% of the pile diameter at twice the design load. This indicates that the available design manual does not provide reasonable parameters for IGM layers. The study was performed to investigate the efficiency of different approaches for determining the design load of bored piles in IGMs. These approaches are statistical, predictions from static pile load tests, numerical, and dynamic wave analysis via a case pile wave analysis program, CAPWAP, a method that calculates friction stresses along the pile shaft. The predicted ultimate capacities range from 5.5 to 10.0 times the pile design capacity. Settlement analysis indicates that the large-diameter pile behaves as a friction pile. The dynamic pile load test results were calibrated relative to the static pile load test. The dynamic load test could be used to validate the pile capacity. Settlement from the dynamic load test has been shown to be about 25% higher than that from the static load test. This can be attributed to the possible development of high pore water pressure in cohesive IGMs. The case study analysis and the parametric study indicate that AASHTO LRFD is conservative in estimating skin friction, tip, and load test resistance factors in IGMs. A new load–settlement response equation for 600 mm to 2000 mm diameter piles and new recommendations for resistance factors φqp, φqs, and φload were proposed to be 0.65, 0.70, and 0.80, respectively.

To investigate the effect of combined end-and-shaft post-grouting on the vertical load-bearing performance of bridge-bored piles in the Dongting Lake area of Hunan, two post-grouted piles were subjected to bi-directional O-cell and top-down load tests before and after combined end-and-shaft grouting, based on the Wushi to Yiyang Expressway project. A comparative analysis was conducted on the bearing capacity, deformation characteristics, and load transfer behavior of the piles before and after grouting. This study also examined the conversion coefficient γ values of different soil layers obtained from the bi-directional O-cell test for bearing capacity calculations. Additionally, the characteristic values of the end bearing capacity, obtained from the bi-directional O-cell and top-down load tests, were compared with the values calculated using the relevant formulas in the current standards, which validated the accuracy of existing regulations and traditional loading methods. The results indicate that the stress distribution along the pile shaft differed between the two test methods. In the bi-directional O-cell test, the side resistance developed from the end to the head, while in the top-down load test, it developed from the head to the end. After combined post-grouting, the ultimate bearing capacity of the piles significantly increased, with side resistance increasing by up to 81.03% and end resistance by up to 105.66%. The conversion coefficients for the side resistance in silty sand and gravel before and after grouting are 0.86 and 0.80 and 0.81 and 0.69, respectively. The characteristic values of the end bearing capacity, as measured by the bi-directional O-cell and top-down load tests, were substantially higher than those calculated using the current highway bridge and culvert standards, showing increases of 133.63% and 86.15%, respectively. These findings suggest that the current standard formulas are overly conservative. Additionally, the measured values from the top-down load test may underestimate the actual bearing capacity of piles in engineering projects. Therefore, it is recommended that future pile foundation designs incorporate both bi-directional O-cell testing and combined post-grouting techniques to optimize design solutions.

Accurate determination of the axial load capacity of the pile is of utmost importance when designing the pile foundation. However, the methods of determining the axial load capacity of the pile in the field are often costly and time-consuming. Therefore, the purpose of this study is to develop a hybrid machine-learning to predict the axial load capacity of the pile. In particular, two powerful optimization algorithms named Herd Optimization (PSO) and Genetic Algorithm (GA) were used to evolve the Random Forest (RF) model architecture. For the research, the data set including 472 results of pile load tests in Ha Nam province—Vietnam was used to build and test the machine-learning models. The data set was divided into training and testing parts with ratio of 80% and 20%, respectively. Various performance indicators, namely absolute mean error (MAE), mean square root error (RMSE), and coefficient of determination (R 2 ) are used to evaluate the performance of RF models. The results showed that, between the two optimization algorithms, GA gave superior performance compared to PSO in finding the best RF model architecture. In addition, the RF-GA model is also compared with the default RF model, the results show that the RF-GA model gives the best performance, with the balance on training and testing set, meaning avoiding the phenomenon of overfitting. The results of the study suggest a potential direction in the development of machine learning models in engineering in general and geotechnical engineering in particular.