Evolutionary Polynomial Regression Research Articles

An EPR model for predicting the bearing capacity of single and double-strip foundations near earth slope crests.

It is imperative to understand how foundations behave on earthen slopes to accurately predict their allowable carrying capacity in geotechnical engineering. A comprehensive finite element (FE) simulation with PLAXIS 2D was conducted to assess the effects of various parameters on the bearing capacity (BC) of single- and double-strip foundations placed near the earth's slope crest. The specified parameters include foundation width (B) and depth (Df/B); setback distance between the slope edge and foundation (b/B); soil internal friction (ϕ) and cohesion (c); slope inclination (β); and spacing between foundations (S/B). In addition, the numerically simulated database was used to develop simple mathematical expressions for predicting the capacities in both cases using evolutionary polynomial regression (EPR). The results revealed that the bearing capacity of single- and double-strip foundations increased with an increase in all studied parameters except slope inclination. For single-strip foundations, the outcomes demonstrated that slope inclination has no impact on BC when it is located 6B from the slope edge. However, under interference conditions, the critical center-to-center spacing between foundations is 3-4B, beyond which they behave as individual foundations. Additionally, EPR provides a robust method of predicting the BC of single- and double-strip foundations within slope crests based on the strong correlation of various statistical criteria between simulated and predicted results from training, validation, and testing. Finally, according to sensitivity analysis, in both single and double-strip foundations resting on an earthen slope crest, b/B, B, and ϕ are the most important input parameters that impact the output results.

Modeling the influence of lime on the unconfined compressive strength of reconstituted graded soil using advanced machine learning approaches for subgrade and liner applications.

In the field of soil mechanics, especially in transportation and environmental geotechnics, the use of machine learning (ML) techniques has emerged as a powerful tool for predicting and understanding the compressive strength behavior of soils especially graded ones. This is to overcome the sophisticated equipment, laboratory space and cost needs utilized in multiple experiments on the treatment of soils for environmental geotechnics systems. This present study explores the application of machine learning (ML) techniques, namely Genetic Programming (GP), Artificial Neural Networks (ANN), Evolutionary Polynomial Regression (EPR), and the Response Surface Methodology in predicting the unconfined compressive strength (UCS) of soil-lime mixtures. This was for purposes of subgrade and landfill liner design and construction. By utilizing input variables such as Gravel, Sand, Silt, Clay, and Lime contents (G, S, M, C, L), the models forecasted the strength values after 7 and 28 days of curing. The accuracy of the developed models was compared, revealing that both ANN and EPR achieved a similar level of accuracy for UCS after 7 days, while the GP model performed slightly lower. The complexity of the formula required for predicting UCS after 28 days resulted in decreased accuracy. The ANN and EPR models achieved accuracies of 85% and 82%, with R2 of 0.947 and 0.923, and average error of 0.15 and 0.18, respectively, while the GP model exhibited a lower accuracy of 66.0%. Conversely, the RSM produced models for the UCS with predicted R2 of more than 98% and 99%, for the 7- and 28- day curing regimes, respectively. The RSM also produced adequate precision in modelling UCS of more than 14% against the standard 7%. All input factors were found to have almost equal importance, except for the lime content (L), which had an average influence. This shows the importance of soil gradation in the design and construction of subgrade and landfill liners. This research further demonstrates the potential of ML techniques for predicting the strength of lime reconstituted G-S-M-C graded soils and provides valuable insights for engineering applications in exact and sustainable subgrade and liner designs, construction and performance monitoring and rehabilitation of the constructed civil engineering infrastructure.

Finite element and evolutionary polynomial regression analyses of the effect of a cavity on the bearing capacity factor $${{\varvec{N}}}_{{\varvec{c}}}$$ of strip footing

Finite element and evolutionary polynomial regression analyses of the effect of a cavity on the bearing capacity factor $${{\varvec{N}}}_{{\varvec{c}}}$$ of strip footing

An optimized prediction of FRP bars in concrete bond strength employing soft computing techniques

The precise estimation of the bonding strength between concrete and fiber-reinforced polymer (FRP) bars holds significant importance for reinforced concrete structures. This study introduces a new methodology that utilizes soft computing methods to enhance the prediction of FRP bars' bonding strength. A significant compilation of experimental bond strength tests is assembled, covering various variables. Significant variables that affect bonding strength are found in the study of this database. The prediction process is optimized using soft computing methods, particularly Gene Expression Programming (GEP) and the Multi-Objective Genetic Algorithm Evolutionary Polynomial Regression (MOGA-EPR).The proposed soft computing approaches accommodate complex relationships and optimize prediction accuracy depending on the input variables. Results demonstrate its effectiveness in predicting bond strength and comparing it with existing codes and other models from the literature. The results have shown that the MOGA-EPR and the GEP models have high R2 values between 0.91 and 0.94. The proposed new models enhance the reliability and efficiency of designing and assessing FRP-reinforced concrete.

Prediction of steel plate-based damper for improving the behavior of concentrically braced frames based on RSM and ML approaches for sustainable structures.

The stiffness (K) and slenderness factor (λ) of a steel plate-based damper has been studied on the basis of elastic-inelastic-plastic buckling (EIP) modes and flexural/shear/flexural-shear failure mechanisms (FSF-S), which has been designed for the improvement of the behavior of concentrically braced frames. Steel plate-based dampers offer significant benefits in terms of mode shapes and failure mechanisms, contributing to improved dynamic performance, enhanced structural resilience, and increased safety of civil engineering structures. Their effectiveness in mitigating dynamic loads makes them a valuable tool for engineers designing structures to withstand extreme environmental conditions and seismic events. This study was undertaken by using the learning abilities of the response surface methodology (RSM), artificial neural network (ANN) and the evolutionary polynomial regression (EPR). Steel plate dampers are special structural designs used to withstand the effect of special loading conditions especially seismic effects. Its design based on the prediction of its stiffness (K) and slenderness factor (λ) cannot be overlooked in the present-day artificial intelligence technology. In this research work, thirty-three entries based on the steel plate damper geometrical properties were recorded and deployed for the intelligent forecast of the fundamental properties (λ and K). Design ratios of the steel plate damper properties were considered and models behavior was recorded. From the outcome of the model, it can be observed that even though the EPR and ANN in that order outclassed the other techniques, the RSM produced model minimization and maximization features of the desirability levels, color factor scales and 3D surface observation, which shows the real model behaviors. Overall, the EPR with R2 of 0.999 and 1.000 for the λ and K, respectively showed to be the decisive model but the RSM has features that can be beneficial to the structural design of the studied steel plate damper for a more robust and sustainable construction. With these performances recorded in this exercise, the techniques have shown their potential to be applied in the prediction of steel damper stiffness with optimized characteristic features to withstand structural stresses.

Machine learning base models to predict the punching shear capacity of posttensioned UHPC flat slabs

The aim of this research is to present correction factors for the punching shear formulas of ACI-318 and EC2 design codes to adopt the punching capacity of post tensioned ultra-high-performance concrete (PT-UHPC) flat slabs. To achieve that goal, the results of previously tested PT-UHPC flat slabs were used to validate the developed finite element method (FEM) model in terms of punching shear capacity. Then, a parametric study was conducted using the validated FEM to generate two databases, each database included concrete compressive strength, strands layout, shear reinforcement capacity and the aspect ratio of the column besides the correction factor (the ratio between the FEM punching capacity and the design code punching capacity). The first considered design code in the first database was ACI-318 and in the second database was EC2. Finally, there different “Machine Learning” (ML) techniques manly “Genetic programming” (GP), “Artificial Neural Network” (ANN) and “Evolutionary Polynomial Regression” (EPR) were applied on the two generated databases to predict the correction factors as functions of the considered parameters. The results of the study indicated that all the developed (ML) models showed almost the same level of accuracy in terms of the punching ultimate load (about 96%) and the ACI-318 correction factor depends mainly on the concrete compressive strength and aspect ratio of the column, while the EC2 correction factor depends mainly on the concrete compressive strength and the shear reinforcement capacity.

Using symbolic machine learning to assess and model substance transport and decay in water distribution networks

Drinking water infrastructures are systems of pipes which are generally networked. They play a crucial role in transporting and delivering clean water to people. The water quality analysis refers to the evaluation of the advective diffusion of any substance in drinking water infrastructures from source nodes. Such substances could be a contamination for the system or planned for the disinfection, e.g., chlorine. The water quality analysis is performed by integrating the differential equation in the pipes network domain using the kinetics of the substance decay and the Lagrangian scheme. The kinetics can be formulated using a specific reaction order depending on the substance characteristics. The basis for the integration is the pipes velocity field calculated by means of hydraulic analysis. The aim of the present work is to discover the intrinsic mechanism of the substance transport in drinking water infrastructures, i.e., their pipes network domain, using the symbolic machine learning, named Evolutionary Polynomial Regression, which provides “synthetic” models (symbolic formulas) from data. We demonstrated, using one real network and two test networks, that the concentration at each node of the network can be predicted using the travel time along the shortest path(s) between the source and each node. Additionally, the formula models provided by symbolic machine learning allowed discovering that a unique formula based on kinetic reaction model structure allows predicting the residual substance concentration at each node, given the source node concentration, surrogating with a good accuracy the integration of the differential equations.

Local scour depth at piles group exposed to regular waves: On the assessment of expressions based on classification concepts and evolutionary algorithms

An accurate estimation of local scour depth around piles group is inevitably essential to provide stability of marine structures. Over the past decades, many investigations have been made to understand the scouring process at piles group exposed to waves for field and experimental scales. This study aims to predict the wave-induced local scour depth by various robust Data Driven Models (DDMs) and Machine Learning Models (MLMs) developed by classification and evolutionary concepts: Model Tree (MT), Evolutionary Polynomial Regression (EPR), Multivariate Adaptive Regression Spline (MARS), and Gene-Expression Programming (GEP). From relevant literature, 125 data series were employed to provide empirical relationships for scour depth predictions. The raw variables were related to bed sediment, pile configuration, pile geometry, approaching flow, and wave characteristics. Non-dimensional parameters have been obtained through the Buckingham theorem in order to control local scour depth. In this way, the ratio of spacing between piles to pile diameter (G/D), sediment number (Ns), pile arrangement number (ratio of the number of piles parallel to the flow; m; to the number of piles normal to the flow; n), Shields parameter (θ), and Keulegan-Carpenter (KC) were considered. From the training and testing stages of Artificial Intelligence Models (AIMs), it was found that regression equation given by MARS model provided better performance (e.g., Correlation Coefficient [R] = 0.9297, Root Mean Square Error [RMSE] = 0.3489, and Scatter Index [SI] = 0.2765) than other AI models’ efficiency. Additionally, the performance of AI models was assessed for various ranges of dimensionless parameters (i.e., G/D, Ns, and θ) versus KC variation. MARS model had the best performance for G/D = 0–1 and KC < 10 (R = 0.9917 and RMSE = 0.2198) and 0.4≤θ < 0.5 and 15 < KC < 25 whereas EPR model had promising efficiency in the its highest level for Ns = 1–3 and 10 < KC ≤ 15 (R = 0.9941 and RMSE = 0.0771) than other AI models. For the sensitivity analysis, the mutation rate and number of chromosomes are examined for the GEP model, the K-fold number and number of parameters for the MARS model, and the number of algebraic terms and number of chromosomes for the EPR model. Furthermore, ranking analysis of AI models indicated that MARS model had the best performance (Ranking Performance Index [RPI] = 0.7348) and followed by GEP (0.3412), M5MT (0.2999), and EPR (0.2848).

New empirical equations to assess energy efficiency of flow-dissipating vortex dropshaft

Vortex structures are generally considered the common energy-dissipating infrastructures in order to convey urban surface waters. During transferring wastewater in such urban networks, the major proportion of water energy is dissipated and additionally, the hydraulic performance of these infrastructures is evaluated on the basis of the dissipation of energy efficiency in the vortex dropshaft. Over the last decade, the literature review of vortex structures demonstrated that there is no reliable mathematical expression to predict energy efficiency although various experimental investigations were done to evaluate the hydraulics of vortex structures. In this research, robust numerical models based on conceptions of Artificial Intelligence (AI) methodologies (i.e., classification, non-parametric analysis, and evolutionary computing) were developed by using 144 experimental data extracted from a physical model of vortex dropshaft. Through analysis of experiments, three main factors, called flow Froude number (Fr), the ratio of sump height (Hs) to shaft diameter (D), and the ratio of drop total height (L) to shaft diameter (D) were determined to estimate the efficiency of flow energy dissipation. The numerical models were calibrated/trained by different algorithms such as multi/single-objective optimization strategies (e.g., Particle Swarm Optimization [PSO] and Multi-Objective Genetic Algorithm [MOGA]). Various statistical tests, known as quantitative assessments for training and testing stages of AI models, demonstrated that Evolutionary Polynomial Regression (EPR) had the best performance (Coefficient of Correlation [R] = 0.9736, Root Mean Square Error [RMSE] = 0.5923, and Mean Absolute Percentage Error [MAPE] = 0.5035) and followed by MT (R = 0.9665, RMSE = 0.7039, and MAPE = 0.6225), GEP (R = 0.9113, RMSE = 1.0803, and MAPE = 0.9535), and MARS (R = 0.8337, RMSE = 1.6473, and MAPE = 1.1010). Additionally, expressions given by AI models provided more accurate predictions of energy dissipation efficiency when compared with a regression-based equation (R = 0.9113, RMSE = 0.7039, MAPE = 0.6225) from the literature. Additionally, variations of three effective factors versus the predicted responses (values of energy dissipation efficiency given by AI models) were in coincidence with experimental observations. Overall, this study proved that the usability of mathematical expressions could provide new insights for experts to assess the hydraulic evaluation of vortex structures before their physical and prototype models are constructed.

Vulnerability of the rip current phenomenon in marine environments using machine learning models

Hidden and perilous rip currents are one of the primary factors leading to drownings of beach swimmers. By identifying the coastal areas with the highest likelihood of generating rip currents, it becomes possible to prevent fatalities and mitigate economic losses associated with these hazardous currents. Rip currents are characterized as streams of water moving towards the open sea, forming within the area where waves break, due to variations in wave-induced radiation stresses and pressure along the coastline. This study utilized nine different Machine Learning (ML) models, including M5 Model Tree (MT), Multivariate Adaptive Regression Spline (MARS), Gene Expression Programming (GEP), Evolutionary Polynomial Regression (EPR), Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Adaptive Boosting (AdaBoost), and Stacked ML models, to estimate the Relative Tide Range (RTR) values for 50 southern beaches in China. Through this approach, we gathered a reliable dataset from prior research conducted on the southern coast of China. In this study, two parameters, namely dimensionless fall velocity parameter (Ω) and tide range (TR) are used to predict the vulnerability of rip current event. The results of the AI models were assessed by various statistical analyses (Correlation of Coefficient [R], Root Mean Square Error [RMSE], violin diagram, heatmap, and taylor diagram) for training and testing stages. Accordingly, the MARS model exhibited superior performance compared to other AI models in accurately predicting the RTR value. The outcomes substantiated the significant effectiveness and capability in estimating the RTR with high accuracy. Southern China coasts have relative high risk level of rip current, necessitating the attention and strategic management of this dangerous current by the beach managers.

The Impact of Shear Reinforcement Amount and Arrangement on the Shear Capacity of Shallow RC Beams: An Experimental Study

This study investigates the impact of shear reinforcement amount and arrangement on the shear capacity of shallow/wide RC beams. Seven specimens of shallow/wide beams with different ultimate shear reinforcement stress (μ.Fys), longitudinal spacing to depth ratio (S/d), and transversal spacing to depth ratio (S’/d) were tested under a monotonic three-point bending test. All the specimens were designed to fail at shearing. The results showed that the shear reinforcement was fully functioning until it yielded; also, the amount of shear reinforcement had the major impact on the shear capacity; in addition, the transverse spacing had more influence on the shear capacity than the longitudinal spacing. The measured shear capacities were compared to six design codes, in which the results ranged from 95% to 110%, with the Japanese code (JSCE) being the closest to the experimental results. Two AI-based predicting equations, “Genetic Programming” (GP) and “Evolutionary Polynomial Regression” (EPR), were also compared to the experimental, with accuracies of 78% and 86% of the measured capacities, respectively. Initial stiffness, final stiffness, dissipated energy, and ductility were all discussed for the seven specimens, with ultimate shear reinforcement stress being the most impactful on the total shear capacity of the wide beams. Doi: 10.28991/CEJ-2023-09-12-013 Full Text: PDF

Shear strength assessment of reinforced recycled aggregate concrete beams without stirrups using soft computing techniques

This paper presents a study to predict the shear strength of reinforced recycled aggregate concrete beams without stirrups using soft computing techniques. The methodology involves the development of a Multi-Objective Genetic Algorithm Evolutionary Polynomial Regression (MOGA-EPR) and Gene Expression Programming (GEP) models. The input variables considered are the longitudinal reinforcement ratio, recycled coarse aggregate ratio, beam cross-section dimensions, and concrete compressive strength. Data collected from the literature were used to train and validate the models. The results showed that the MOGA-EPR and GEP models can accurately predict the shear strength of beams without stirrups. The models also performed better than equations from the codes and literature. This study provides an alternative approach to accurately predict the shear strength of reinforced recycled aggregate concrete beams without stirrups.

An evolutionary-based polynomial regression modeling approach to predicting discharge flow rate under sheet piles

The discharge flow rate beneath sheet plies is an essential parameter in designing these water retaining structures. This paper presents a unified framework for modeling and predicting discharge flow rate using an evolutionary-based polynomial regression technique. EPR (Evolutionary Polynomial Regression) is a data-driven method based on evolutionary computing to search for polynomial structures representing a system. The input parameters in the modeling procedure included the sheet pile height, upstream water head, and the hydraulic conductivity anisotropy ratio. Due to ever-increasing demand for water, a widely held view on predicting and controlling the available water behind reservoirs, dams, barrages, and weirs is of vital importance. To this end, the sheer novelty of the current study has been worn off through the development of a comprehensive model to predict the flow rate considering the most effective variables in the seepage issue. To the best of our knowledge, the research conducted in the literature has yet to cover the whole seepage problem using a comprehensive database extracted by numerical methods; thus, a comprehensive finite-element-based artificial database including 1000 data lines was created using the Scaled Boundary Finite Element Method (SBFEM) by simulating seepage beneath sheet plies covering a considerably wide range of seepage-related real-world values. The database was then employed to develop and validate the EPR flow rate prediction model. Data were divided into training (used for creating the models) and testing (for validating the developed models) data based on a statistical process. The procedure for preparing the data and developing and validating the models is presented in detail in this paper. The main advantage of the proposed models over a conventional and neural network and most GP (Genetic Programming)-based constitutive models is that they provide the optimum structure for the material constitutive model representation as well as its parameters, directly from raw experimental (or field) data. EPR can learn nonlinear and complex material behavior without any prior assumptions on the constitutive relationships. The proposed algorithm captures and transparently presents relationships between contributing parameters in polynomial expressions providing the user with a clear insight into the problem. EPR-based model predictions demonstrated an excellent agreement with the unseen simulated data used for validating the developed model. A parametric study on the presented models was conducted to investigate the effects of the contributing parameters on model predictions and the consistency of the parameter relationships with the database. Results of the parametric study showed that the effects of variations in the contributing parameters on EPR predictions are in line with the expected behavior. The merits and advantages of the proposed technique are discussed in the paper.

A performance-based hybrid deep learning model for predicting TBM advance rate using Attention-ResNet-LSTM

The technology of tunnel boring machine (TBM) has been widely applied for underground construction worldwide; however, how to ensure the TBM tunneling process safe and efficient remains a major concern. Advance rate is a key parameter of TBM operation and reflects the TBM-ground interaction, for which a reliable prediction helps optimize the TBM performance. Here, we develop a hybrid neural network model, called Attention-ResNet-LSTM, for accurate prediction of the TBM advance rate. A database including geological properties and TBM operational parameters from the Yangtze River Natural Gas Pipeline Project is used to train and test this deep learning model. The evolutionary polynomial regression method is adopted to aid the selection of input parameters. The results of numerical experiments show that our Attention-ResNet-LSTM model outperforms other commonly-used intelligent models with a lower root mean square error and a lower mean absolute percentage error. Further, parametric analyses are conducted to explore the effects of the sequence length of historical data and the model architecture on the prediction accuracy. A correlation analysis between the input and output parameters is also implemented to provide guidance for adjusting relevant TBM operational parameters. The performance of our hybrid intelligent model is demonstrated in a case study of TBM tunneling through a complex ground with variable strata. Finally, data collected from the Baimang River Tunnel Project in Shenzhen of China are used to further test the generalization of our model. The results indicate that, compared to the conventional ResNet-LSTM model, our model has a better predictive capability for scenarios with unknown datasets due to its self-adaptive characteristic.

Soft computing models for assessing bond performance of reinforcing bars in concrete at high temperatures

The bond between steel and concrete in reinforced concrete structures is a multifaceted and intricate phenomenon that plays a vital role in the design and overall performance of such structures. It refers to the adhesion and mechanical interlock between the steel reinforcement bars and the surrounding concrete matrix. Under elevated temperatures, the bond is more complex under higher temperatures, yet having an accurate estimate is an important factor in design. Therefore, this paper focuses on using data-driven models to explore the performance of the concrete-steel bond under high temperatures using a Gene Expression Programming (GEP) soft computing model. The GEP models are developed to simulate the bond performance in order to understand the effect of high temperatures on the concrete-steel bond. The results were compared to the multi-objective evolutionary polynomial regression analysis (MOGA-EPR) models for different input variables. The new model would help the designers with strength predictions of the bond in fire. The dataset used for the model was obtained from experiments conducted in a laboratory setting that gathered a 316-point database to investigate concrete bond strength at a range of temperatures and with different fibre contents. This study also investigates the impact of the different variables on the equation using sensitivity analysis. The results show that the GEP models are able to predict bond performance with different input variables accurately. This study provides a useful tool for engineers to better understand the concrete-steel bond behaviour under high temperatures and predict concrete-steel bond performance under high temperatures.

Application of soft computing in predicting the compressive strength of self-compacted concrete containing recyclable aggregate

AbstractSelf-compacting concrete (SCC) is a type of concrete known for its environmental benefits and improved workability. In this study, data-driven approaches were used to anticipate the compressive strength (CS) of self-compacting concrete (SCC) containing recycled plastic aggregates (RPA). A database of 400 experimental data sets was used to assess the capabilities of multi-objective genetic algorithm evolutionary polynomial regression (MOGA-EPR) and gene expression programming (GEP). The analysis results indicated that the proposed equations provided more accurate CS predictions than traditional approaches such as the linear regression model (LRM). The proposed equations achieved lower mean absolute error (MAE) and root mean square error (RMSE) values, a mean close to the optimum value (1.0), and a higher coefficient of determination (R2) than the LRM. As such, the proposed approaches can be utilized to obtain more reliable design calculations and better predictions of CS in SCC incorporating RPA.

AI-Based Estimation of Swelling Stress for Soils in South Africa

AbstractSoil swelling is one of the major causes of structural failure, due to excessive moisture saturation and desaturation moisture. In situ measurement of swelling stress is somewhat impossible and requires tedious routine site observation. The use of artificial intelligence to predict the swelling stress of in situ soil is highly recommended, because of the complex behavior of soil upon moisture absorption. Because of this challenge, this study is channeled towards the prediction of swelling stress using basic geotechnical properties to save the challenges of repeating geotechnical experimental tests. In this study, the swelling stress of soils collected from 15 locations in 5 sites across South Africa has been predicted by using the artificial neural network (ANN), genetic programming (GP), and evolutionary polynomial regression (EPR)-based intelligent techniques. Multiple data were collected through laboratory experiments on the predictors: gravimetric moisture content (GMC), plasticity index (Ip), dry density (γd), free swell index (FSI), degree of saturation (S), matric suction (Ψm) and the target, and swelling stress (Psm). This predictive model was aimed at proposing models, which will help earthwork designers and constructors in South Africa overcome the rampant visit to the laboratory in search of soil data needed for geotechnical engineering designs. The soils showed their potential for swelling, which was eventually confirmed by the sensitivity analysis of the intelligent models. The performance indices of the models showed that ANN outclassed the other techniques with a performance accuracy of 93.6% at an error of 1.9%. Also, the sensitivity analysis showed that the plasticity index and matric suction were the most influential to the models. With the predicted models, future earthworks in South Africa can quickly forecast swelling stress prior to designs and construction more so in hydraulically bound environments.

EVOLUTIONARY POLYNOMIAL REGRESSION FOR PREDICTING THE UNCONFINED COMPRESSIVE STRENGTH OF LIME-STABILIZED

Peat has a lack of strength, making it unsuitable for foundations or road construction. The results of 104 unconfined compressive strength (UCS) on peat stabilized with various lime percentages are presented in this article. The strength performance of peat soils stabilized with lime in soaked and unsoaked conditions was also evaluated. Evolutionary Polynomial Regression (EPR) models are constructed to predict the UCS values using the data obtained from the tests. After evaluating the various functions, the EPR model with a hyperbolic tangent function is the best model for predicting the UCS. The proposed EPR models may benefit practitioners when determining the optimal amount of stabilizers to utilize in soft-ground improvement projects. In this section, we show and expand on applying the suggested models to the design of soft ground mixing. Additionally, by applying the created relationships to a new instance, the comprehensiveness of the established relationships is shown, as is the practicality of the suggested relationships for usage during the preliminary design stage. The findings of this study have important implications for estimating the quantity of lime required in ground peat improvement work.

Prediction of the mechanical performance of compressed circular CFST columns with circumferential debonding gap

Concrete-filled steel tubular (CFST) structures have been widely used in high-rise buildings and long-span bridges for the high strength and ductility due to the confinement effect of the concrete core provided by the steel tube. This beneficial composite action could be damaged by circumferential debonding gap (CDG), which is nearly unavoidable in real CFST structures. Reliable formulations to predict the influence of CDG on the mechanical performance of compressed circular CFST columns are needed. In this paper, the experimental parameters and results of compressed circular CFST columns with CDG are summarized to discuss the deficiencies of the parameters chosen in the tests. Detailed finite element simulation procedures for compressed circular CFST columns with CDG, based on ABAQUS, are then introduced, verifying their accuracy by test results. Based on the validated finite element models, a database of compressed circular CFST columns with and without CDG, including strength, ductility and stiffness of specimens is established; it is obtained by covering suitable parameter ranges for CDG and CFST structures, in accordance with real applications. Successively, new formulations able to predict the ultimate load capacity, the reduction factor, the ductility and the stiffness of specimens are derived by the Evolutionary Polynomial Regression (EPR) methodology, demonstrating good accuracy and low complexity.

Using FEM-AI Technique to Predict the Behavior of Strip Footing Rested on Undrained Clay Layer Improved with Replacement and Geo-Grid

The objective of this research is to predict how strip footings behave when rested on an undrained clay layer enhanced using a top replacement layer with and without a geo-grid. The study was conducted in several stages, including collecting load-settlement curves from "Finite Element Method" (FEM) models with different clay strengths, replacement thicknesses, and axial stiffnesses of the geo-grid. These curves were then idealized using a hyperbolic model, and the idealized hyperbolic parameters were predicted using three different AI techniques. According to the numerical results, the ultimate bearing pressure of pure clay models was found to be five times the undrained strength of the clay. These findings align with most established empirical bearing capacity formulas for undrained clays. The results also suggest that the initial modulus of the subgrade reaction is solely influenced by replacement thickness. Additionally, the enhancement in subgrade reaction due to the replacement layer decreases with increasing clay strength. However, the percentage of improvement decreased with higher clay strength. Moreover, the impact of the geo-grid was significant for settlement beyond 50mm, and it was more impactful in soft clay than in stiff clay. Finally, the research proposed predictive models employing the "Genetic Programming" (GP), "Artificial Neural Networks" (ANN), and "Evolutionary Polynomial Regression" (EPR) techniques, and these models exhibited an accuracy of about 88%. Doi: 10.28991/CEJ-2023-09-05-014 Full Text: PDF