No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Shear Resistance Prediction of Concrete Beams Reinforced by FRP Bars Using Artificial Neural Networks
Abstract
In the past, remarkable behavior evaluations were carried out on concrete beams reinforced with FRP bars in the longitudinal direction without shear reinforcement. The aim of this study is to develop an artificial neural network (ANN) approach for predicting shear resistance of concrete beams. Proposed method considers geometric and mechanical properties of cross section and FRP bars, and shear span-depth ratio. Capability of the proposed method was compared with existing approaches in the literature using comprehensive database. The existing approaches include the American Concrete Institute design guide (ACI 440.1R-06), ISIS Canadian design manual (ISIS-M03-07), the British Institution of Structural Engineers guidelines (BISE), JSCE Design Recommendation, CNR-DT 203-06 Task Group, and Kara. The findings show that proposed method has excellent agreement with the experimental database.
... However, their inherent capability to effectively identify and replicate patterns renders them an ideal tool for undertaking such a task [23]. In the context of FRP strengthening of RC beams, these models assume a highly specialized role since they possess the capability to address intricate issues that prove challenging to resolve by analytical methods [24]. This is particularly true in cases when there is a lack of fundamental physical or mathematical models to guide the resolution of these challenges. ...
... While there exist alternative training processes, the backpropagation (BP) algorithm is generally known to produce satisfactory outcomes [30]. The Levenberg-Marquardt method is often employed as a post-release training algorithm [24]. Prior to the training phase, the acquired data, consisting of inputs and targets, undergoes normalization or scaling based on known relationships. ...
... In accordance with the observed behavior of the analyses performed using neural networks, we suggest an optimization strategy to generate simple and accurate flexure design equations for NSM-FRP flexure reinforced RC beams. The optimization strategy is created in GMDH [24] due to the fact that GMDH algorithms are specifically suitable for handling multi objective issues. ...
In recent years, there has been a notable increase in the application of near-surface mounted fiber-reinforced polymer (FRP) reinforcement in reinforced concrete structures. Nevertheless, there is a discernible disparity in the accessibility of accurate and customize measures for augmenting flexural strength through the use of near-surface mounted (NSM) reinforcement techniques. Although several basic models have been proposed to predict the flexural capacity achievable with this technology, established codes have not yet provided mathematical equations for this specific purpose. This study presents two separate methodologies with the objective of enhancing the development of suitable code provisions. In the first stage, A comprehensive and reliable database has been developed to leverage the predictive accuracy of neural networks in the computation of the flexural capacity of reinforced beams that utilize near-surface mounted reinforcement. Following this, the results obtained from the neural network are employed to construct a linear equation using the group method of data handling (GMDH) technique. The presented equation has been carefully formulated to produce a concise and simple mathematical expression that enables the determination of the flexural strength of a beam on the field. The evaluation of the accuracy and effectiveness of both the neural network and the suggested equation is conducted in accordance with the requirements specified in ACI 440.R2 for externally bonded reinforcements. The neural network's prediction has a mean absolute error of just 5% in comparison to the experimental results and the GMDH equations exhibit a noteworthy level of concurrence with the experimental outcomes, as they display a mean absolute error of 16%.
... In most cases, the shear failure of FRP-RC beams has been investigated with a similar approach to that used for steel-RC (Nehdi et al. 2007). The shear strength equations that have been proposed in various codes, standards, and guidelines are highly conservative, tending toward over-design and demand a large amount of FRP; therefore, making FRP applications less cost-effective (El-Sayed et al. 2006;Naderpour et al. 2018;Alguhi and Tomlinson 2021). ...
... Existing empirical or semiempirical models for shear strength prediction of FRP-RC members have been developed by simply modifying the methods that are available for steel-RC members and they are based on limited test data (Nehdi et al. 2006(Nehdi et al. , 2007Hoult et al. 2008;Kara et al. 2013;Lee and Lee 2014;Shahnewaz et al. 2016;Naderpour et al. 2018;Alam and Gazder 2020). This resulted in a large scatter in the predictions compared with the experimental results and tended to be conservative. ...
... However, ρ f and a/d were the two less essential features in the SVR model with small relative importance (i.e., approximately 3%). Similar findings were reported by Nehdi et al. (2007), Naderpour et al. (2018), and Alguhi and Tomlinson (2021) through sensitivity analyses. ...
Estimating the shear strength of a fiber-reinforced polymer (FRP)–reinforced-concrete (RC) beam is a complex task that depends on multiple design variables. The use of FRP bars has emerged as a promising alternative to diminish the corrosion problems that are associated with steel reinforcement in adverse environments; however, an accurate and reliable method of shear strength prediction is needed to ensure the economical use of materials and robust designs. Several optimized design equations are available in the literature; however, when utilizing these equations a substantial difference is observed between the predicted outcome (Vpred) and the experimental shear strength (Vexp) result. Therefore, this paper presented a novel approach toward implementing machine learning (ML) algorithms to accurately estimate the shear strength of FRP–RC beams. A large database that consisted of 302 shear test results on FRP-reinforced slender concrete beams without stirrup was collected from the literature to formulate the most efficient prediction model. The performance of each ML algorithm model was compared with the existing design provisions and models. The model interpretation was performed through feature importance analysis to explain the model output compared with a black box. The proposed data-driven ML models demonstrated a high level of accuracy and excellent performance and were superior to the existing shear strength models. In addition, a simple graphical user interface (GUI) was developed to aid practicing engineers when estimating shear strength without the need for complicated design procedures.
... Neural network modeling is inspired by the biological function of the human brain [14]. These networks are an information processing tools [15] and are influential in solving complex civil engineering problems [39]. General computational neurons are illustrated in Fig. 2. It has a computational neuron, an input, a nerve cell, and an output like biological neurons. ...
... Eqs. (5) and (6) calculate the main criteria for stopping training [39]. ...
... Also, 20% of the database was considered for data validation to control network over-training. Data validation errors start to increase when overtraining happens, so an over-trained ANN model has poor predictive performance [39]. Finally, the remaining 20% of the database was assigned to measure the network performance after training. ...
Single fiber pull-out and fiber-matrix interfacial interaction play an essential role in understanding the mechanical behavior of fiber-reinforced cementitious composites. The present study introduces a computational model for predicting the maximum fiber pull-out force and corresponding bond slip. An extensive literature survey was performed to create a pertinent comprehensive experimental database. A total of 382 experimental data were utilized to develop and train the Artificial Neural Network (ANN) models. The model input parameters included the fiber embedded length, fiber inclination angle, fiber tensile strength, fiber length-to-diameter ratio, loading rate, water-to-cement ratio, concrete compressive strength, and fiber geometry. The model output consisted of the maximum pull-out force and corresponding slip. The results indicate that ANN with two hidden layers and 12 neurons was adequate for predicting the outputs with a mean absolute percentage error (MAPE) of less than 10%. To obtain the importance of the inputs on the outputs (the maximum fiber pull-out force and the corresponding slip), a sensitivity analysis was done based on the Milne formula on the proposed ANN. According to the results, it was found that among the eight inputs, the parameters of the geometric shape of the fibers (straight, hooked-end and spiral fibers) and fiber tensile strength have the highest effect on the outputs, with an impact percentage of 16.1 and 15.1, respectively. The mean square error (MSE) was 0.9 for the maximum pull-out force and 0.14 for slip, respectively. Overall, the proposed executed model attained reasonable predictions and could offer a data driven approach to optimizing fiber-reinforced cementitious composites.
... Naderpour, et al. [41] created an artificial neural network (ANN) method to predict concrete beam shear strength. The suggested method takes into account the shear span-depth ratio as well as the geometric and mechanical features of cross sections and FRP bars. ...
... Naderpour, et al. [41] Created an artificial neural network (ANN) method to predict concrete beam shear strength. ...
... Naderpour and Alavi [28] provided a fuzzy-based model for predicting the shear contribution of FRP in RC beams strengthened by externally bonded FRP sheets. Furthermore, several studies have been performed to investigate the shear behavior of FRP-RC beams using ML-based techniques [32][33][34][35][36][37]. In 2011, Kara [32] utilized gene expression programming (GEP) to obtain a shear capacity prediction model for FRP-RC beams without stirrups, where it was shown that the GEP model performed better as compared to the available shear design guidelines. ...
... In 2011, Kara [32] utilized gene expression programming (GEP) to obtain a shear capacity prediction model for FRP-RC beams without stirrups, where it was shown that the GEP model performed better as compared to the available shear design guidelines. Furthermore, Lee and Lee [33], Jumaa and Yousif [34], and Naderpour et al. [35] Polymers 2023, 15, 2857 3 of 20 developed ANN models to evaluate the shear capacity of FRP-RC beams without stirrups. Moreover, Golafshani and Ashour [36] proposed a new model using biogeography-based programming (BBP) to predict the shear capacity of FRP-RC beams without stirrups based on an experimental database of 138 test specimens. ...
The shear strength prediction of concrete beams reinforced with FRP rebars and stirrups is one of the most complicated issues in structural engineering applications. Numerous experimental and theoretical studies have been conducted to establish a relationship between the shear capacity and the design variables. However, existing semi-empirical models fail to deliver precise predictions due to the intricate nature of shear mechanisms. To provide a more accurate and reliable model, machine learning (ML) techniques are adopted to study the shear behavior of concrete beams reinforced with FRP rebars and stirrups. A database consisting of 120 tested specimens is compiled from the reported literature. An artificial neural network (ANN) and a combination of ANN with a genetic optimization algorithm (GA-ANN) are implemented for the development of an ML model. Through neural interpretation diagrams (NID), the critical design factors, i.e., beam width and effective depth, shear span-to-depth ratio, compressive strength of concrete, FRP longitudinal reinforcement ratio, FRP shear reinforcement ratio, and elastic modulus of FRP longitudinal reinforcement rebars and FRP stirrups, are identified and determined as input parameters of the models. The accuracy of the proposed models has been verified by comparing the model predictions with the available test results. The application of the GA-ANN model provides better statistical results (mean value Vexp/Vpre equal to 0.99, R2 of 0.91, and RMSE of 22.6 kN) and outperforms CSA S806-12 predictions by improving the R2 value by 18.2% and the RMSE value by 52.5%. Furthermore, special attention is paid to the coupling effects of design parameters on shear capacity, which has not been reasonably considered in the models in the literature and available design guidelines. Finally, an ML-regression equation considering the coupling effects is developed based on the data-driven regression analysis method. The analytical results revealed that the prediction agrees with the test results with reasonable accuracy, and the model can be effectively applied in the prediction of shear capacity of concrete beams reinforced with FRP bars and stirrups.
... The rapid development of advanced computerised systems has led to efficient and reliable methods for predicting the structural behaviour of RC members. In recent years, artificial neural networks (ANNs) have been increasingly used in the literature for prediction of shear behaviour (Asteris et al., 2019;Jumaa and Yousif, 2018;Naderpour et al., 2018aNaderpour et al., , 2018b, flexural behaviour (Congro et al., 2021;Kaczmarek and Szymańska, 2016;Kang et al., 2021) and compressive strength of concrete (Asteris and Mokos, 2020;Naderpour et al., 2018aNaderpour et al., , 2018bNikoo et al., 2015). Applications of machine learning methods are also extended to structural members made from steel (Dai et al., 2022;Fang et al., 2021aFang et al., , 2021bRabi et al., 2023), stainless steel (Fang et al., 2021c) and aluminium (Fang et al., 2022a(Fang et al., , 2022b(Fang et al., , 2022c. ...
... The rapid development of advanced computerised systems has led to efficient and reliable methods for predicting the structural behaviour of RC members. In recent years, artificial neural networks (ANNs) have been increasingly used in the literature for prediction of shear behaviour (Asteris et al., 2019;Jumaa and Yousif, 2018;Naderpour et al., 2018aNaderpour et al., , 2018b, flexural behaviour (Congro et al., 2021;Kaczmarek and Szymańska, 2016;Kang et al., 2021) and compressive strength of concrete (Asteris and Mokos, 2020;Naderpour et al., 2018aNaderpour et al., , 2018bNikoo et al., 2015). Applications of machine learning methods are also extended to structural members made from steel (Dai et al., 2022;Fang et al., 2021aFang et al., , 2021bRabi et al., 2023), stainless steel (Fang et al., 2021c) and aluminium (Fang et al., 2022a(Fang et al., , 2022b(Fang et al., , 2022c. ...
Stainless steel reinforcement has becoming increasingly popular in the construction industry in recent years owing mainly to its distinctive characteristics and excellent mechanical properties. There is a real need to develop a fundamental understanding of the bond behaviour of stainless steel reinforced concrete. This paper investigates the bond behaviour of stainless steel reinforced concrete using the advancement of the artificial neural networks and compares the performance to experimental data available in the literature with reference to existing bond design rules in international design standards. Accordingly, a new bond design formula is proposed to predict the bond strength capacity of stainless steel reinforcement. The results show an excellent agreement between the experimental results and the predictions of the ANN model. Both Eurocode 2 and model code 2010 are shown to be extremely conservative compared with ANN predictions. The proposed ANN-based formula provides an excellent basis for engineers to specify bond strength of stainless steel reinforcement in RC members in an efficient and sustainable manner, with minimal wastage of materials.
... The research studies [44][45][46][47][48][49][50][51][52] in civil engineering highlight machine learning's potential to enhance and broaden traditional experimental research methodologies. However, the limited sample sizes and relatively simple models used in these studies restricted their ability to fully leverage the available data. ...
Fiber-reinforced polymer (FRP) bars have better tensile strength and durability than normal steel bars. The use of FRP bars to replace steel bars in traditional concrete structures has attracted much attention in recent years. Similar to steel bars, the bond strength between FRP bars and concrete is a very important parameter in FRP-reinforced concrete structures, which directly affects the bearing capacity and safety of the structure. At present, the accuracy of calculating the bond strength between FRP bars and concrete under the action of high temperature still needs to be improved. In this study, a prediction model of bond strength between FRP bars and concrete at high temperatures was constructed by collecting 151 sets of experimental data, and six machine learning algorithms were used to construct the prediction model and perform parameter importance and sensitivity analyses. The results show that the XGBoost model can predict the bond strength between FRP bars and concrete more accurately. The conclusions obtained can provide a reference for the design of the specification to a certain extent.
... In the recent two decades, experts have heavily favored data-driven techniques, particularly artificial neural network (ANN) [33,34], to predict desirable outcomes, in particular about the mechanical strength of GPC [35][36][37][38][39][40]. ANN is a computational model that simulates the functioning of the human brain and is useful for resolving complex engineering problems [41]. ...
Geopolymer concrete (GPC) is an efficient alternative to traditional construction materials, utilizing amorphous or semicrystalline waste–based binders. It enhances durability than Portland cement, remaining environment friendly and cost-effective, while promoting waste reuse. Despite extensive use and research, predicting concrete strength based on the mixed chemical composition of constituents remains challenging due to consideration of limited variables, small dataset sizes, and traditional models struggling with nonlinearity. The primary objective is to develop an artificial neural network (ANN) model to enhance the accuracy of compressive strength (C-S) predictions of GPC, which is crucial for the sustainable use of waste-based binders in construction. For this, the study incorporates a comprehensive dataset of 1018 experimental data points of binary and ternary GPC from 43 research sources addressing these issues. Thirteen input parameters, including three major geopolymer binder’s oxide compositions (viz, SiO2, CaO, and Al2O3), are considered to generalize the model across various binder types. With all these, an ANN model was developed to handle the inherent nonlinearity of input parameters, validated through fivefold cross-validation. The relationship between individual parameters and C-S is analyzed. Additionally, regression models (viz., linear regression (LR), Lasso regression, Ridge regression, and XGBoost regressor) were developed to thoroughly assess the performance of the ANN model in comparison to the conventional regression model. The ANN model, superior to other regression techniques, excels in recognizing nonlinear correlations between features and an objective variable, a challenge for traditional regression models. Various metrics assess model performance (e.g., mean absolute error (MAE), root mean square error (RMSE), mean square error (MSE), R-squared (R²)), revealing the ANN model’s superiority over regression models. The ANN model’s proficiency in nonlinear correlations contributes to its effectiveness, yielding low MAE ( = 2.58), RMSE ( = 4.13), and MSE ( = 17.86) values, and high (R² = 0.93) value, signifying minimal deviation between predicted and actual C-S values. By focusing on the oxide composition of binders, the model is more generalized, making it applicable across different binder types and not limited to specific materials. Also, the parametric analysis confirmed the ANN’s ability to capture the effects of input parameters, offering a comprehensive prediction tool. This study highlights the potential of employing the ANN model for predicting material behavior, offering a resource-efficient approach, and showcasing the viability of mixed GPC for sustainable industrial waste utilization. Future research directions include exploring the model’s application under varying environmental conditions and expanding the dataset to enhance its diversity and representativeness.
... In addition, the performance of the proposed method is also compared to that of Random Forest Regression (RFR) [51] and Levenberg-Marquardt Artificial Neural Network (LM-ANN) [52]. RFR and LM-ANN have been demonstrated to be capable tools for modeling the mechanical properties of concrete in various works [53][54][55]. In this study, RFR is implemented with the Scikit-learn library [56] and the LM-ANN model is built within the MATLAB environment with the neural network toolbox [57]. ...
The estimation of lateral strength in reinforced concrete (RC) columns subjected to cyclic loads is crucial in structural design. The failure of RC columns under lateral forces can lead to catastrophic structural collapses. This fact emphasizes the need for accurate assessments of their dynamic behavior. This paper proposes a data-driven model for estimating the lateral strength of RC columns. A historical dataset comprising 12 predictor variables and 247 samples has been compiled to train and validate of the proposed approach. The extreme gradient boosting machine (XGBoost) is employed to establish a predictive relationship between the lateral strength of RC columns and their influencing factors. Since model selection is critical for constructing a reliable prediction mode, this study relies on utilizing metaheuristic approaches, including Genetic Algorithm, Particle Swarm Optimization, Artificial Bee Colony, and Ant Colony Optimization, to optimize the performance of the XGBoost model. Experimental results show that the integration of Ant Colony Optimization and XGBoost can help attain outstanding prediction accuracy with a correlation of determination (R2) of 0.95. Additionally, an asymmetric squared error loss function is utilized to reduce overestimations by 12%. The newly developed method can be utilized in practical applications where reliable predictions of the lateral strength of RC columns under cyclic loads are required.
... They noted that ANN has the potential to predict compressive strength at different early and old ages. Duan et al. [38] assessed different mechanical properties of RAC by various inputs using a back-propagation neural network (BPNN), reporting the viability of this approach in predicting the RAC specifications. Naderpour et al. [39] studied RAC via BPNN with six input parameters, evaluating the effect of each variable on the compressive strength. ...
The mechanical and durability characteristics of concrete are crucial for designing and evaluating concrete structures throughout their entire operational lifespan. The main objective of this research is to use the deep learning (DL) method along with an artificial neural network (ANN) to predict the chloride migration coefficient and concrete compressive strength. An expansive experimental database of nearly 1100 data points was gathered from existing scientific literature. Four forecast models were created, utilizing between 10 and 12 input features. The ANN was used to address the missing data gaps in the literature. A comprehensive pre-processing approach was then implemented to identify outliers and encode data attributes. The use of mean absolute error (MAE) as an evaluation metric for regression tasks and the employment of a confusion matrix for classification tasks were found to produce accurate results. Additionally, both the compressive strength and chloride migration coefficient exhibit a high level of accuracy, above 0.85, in both regression and classification tasks. Moreover, a user-friendly web application was successfully developed in the present study using the Python programming language, improving the ability to integrate smoothly with the user’s device.
... Reference [73] used the ANN method to assess the compressive strength of concrete using RA (coarse aggregate only) with 14 various inputs and found that this method may be a helpful tool for forecasting RAC compressive strength. In addition, reference [74] used the ANN structure with 6 inputs and 18 hidden nodes to assess the compressive strength of concrete, including construction debris (coarse-grained alternative), as well as the influence of each input factor. The findings showed that water absorption has the most considerable influence on the strength when compared to the other inputs. ...
Recycled aggregate concrete (RAC) is commonly used to lessen the environmental effect of concrete building and demolition waste. The compressive strength of the RAC is one of the most critical factors influencing concrete quality. The compressive strength is assessed by a compression test, which takes a large number of materials and is expensive and time-consuming. With the development of novel concrete mixes and applications, academics are obliged to seek accurate models for forecasting mechanical strength. A significant source of difficulty in compressive strength modeling is that there are many mixture components and testing conditions whose variation significantly influences the predicted values. To this end, this study explores the mixture design of sustainable concrete in order to generate eco-friendly concrete mixes. Tests are conducted on 18 different mixtures comprising different proportions of waste tires, plastic, cement, and red brick to experiment with new green RAC mixtures. For the modeling part, the deep residual neural networks (DRNNs) method is first presented to the problem, aided by a database from the literature for a pretraining task. The proposed DRNNs structure uses shortcuts (i.e., residual connections) that bypass some layers in the deep network structure to alleviate the problem of training with high accuracy. The performance of the proposed DRNNs is evaluated using different goodness of fit measures and compared with well-known machine learning tools. The findings showed that the suggested model could provide credible predictions about the desired mechanical parameter, saving the required lab efforts by 40 %. Finally, a variance-based global sensitivity analysis is performed with the Latin hypercube simulation method to help rank/prioritize each mixture component’s impact on determining the compressive strength in practice while mitigating the potential misrepresentation of results due to the correlations between the input parameters. The analysis showed that cement and waste contents are the most significant ones in their f irst and total order effects.
... The relationship equation between the compressive strength of RC, the content of rubber, and particle size was derived based on data fitting. Meanwhile, a neural network learning method [46][47][48] was used to collect 197 sets of data from 23 studies, and a neural network model of particle size, replacement content, control concrete strength, and rubberized concrete strength was established to predict the static strength of RC. In the study of dynamic impact, the stress-strain curves sets of RC with different rubber admixtures at three different strain rates were analyzed using the SHPB test. ...
Rubberized concrete (RC) has received widespread attention due to its energy absorption and crack resistance properties. However, due to its low compressive strength, it is not recommended for structural applications. The rubber size and content affect RC’s mechanical properties. This study investigated and formulated the behavior of RC with different particle sizes and contents under dynamic and static loading. Quasi-static compressive and dynamic tests were conducted on RC with varying content of rubber (0–30%) and rubber sizes (0.1–20 mm). It was found that the rubber particle size was 0.5mm and the rubber content was 2%. An equation was derived from the experimental data to forecast the impact of rubber size and content on compressive strength. Additionally, by combining the literature and this research’s data, a model was established based on neural networks to predict the strength of RC. SHPB tests were carried out to study the stress–strain curves under dynamic load. The peak stress, fragment analysis, and energy absorption of RC with varying content of rubber and rubber sizes at three different strain rates (100 s⁻¹, 160 s⁻¹, and 290 s⁻¹) were investigated. Equations describing the relationship between dynamic increase factor (DIF), rubber material content, and strain rate on different particle sizes were obtained by fitting. The DIF increased as the content of the rubber increased. By analyzing energy absorption data, it was found that the optimal ratio for energy absorption was RC-0.5-30, RC-0.1-30, and RC-10-30 at strain rates of 100 s⁻¹, 160 s⁻¹, and 290 s⁻¹. This study could be a good guideline for other researchers to easily select the content and size of the rubber in RC for their applications. It also has a positive significance in promoting the development of green building materials.
... The equations representing these evaluation metrics are provided in the equations section (Supplemental Information File Seq. S1-S4) [38][39][40][41]. ...
Chloride ion corrosion has been considered to be one of the main reasons for durability deterioration of reinforced concrete structures in marine or chlorine-containing deicing salt environments. This paper studies the relationship between the amount of fly ash and the durability of concrete, especially the resistance to chloride ion erosion. The heat trend map of total chloride ion factor correlation displayed that the ranking of factor correlations was as follows: sampling depth > cement dosage > fly ash dosage. In order to verify the effect of fly ash dosage on chloride ion resistance, three different machine learning algorithms (RF, GBR, DT) are employed to predict the total chloride content of fly ash proportioned concrete with varying admixture ratios, which are evaluated based on R2, MSE, RMSE, and MAE. The results predicted by the RF model show that the threshold of fly ash admixture in chlorinated salt environments is 30–40%. Replacing part of cement with fly ash in the mixture of concrete below this threshold of fly ash, it could change the phase structure and pore structure, which could improve the permeability of fly ash concrete and reduce the content of free chloride ions in the system. Machine learning modeling using sample data can accurately predict concrete properties, which effectively reduce engineering tests. The development of machine learning models is essential for the decarbonization and intelligence of engineering.
... Due to inference and erudition capabilities, ANNs are widely used in areas where prediction, estimation, pattern recognition, and optimization are needed [30][31][32]. ANNs are being used in the structural engineering field widely, including estimation of structural responses [28][29][30][31][32][33][34][35][36][37][38][39][40][41][42], engineering design [43][44][45][46][47], system identification [48,49], structural condition assessment and monitoring [50][51][52][53][54], and load rating of bridges [55,56]. ...
Splicing steel beams is a widely adopted practice, necessitated by span constraints that arise from challenges in transporting and handling longer beams, which may jeopardize the safety of the onsite workforce. The spliced joint must possess adequate strength and stiffness while effectively distributing the design loads without compromising the structural stability of the system. In this paper, an experimental study was carried out on various spliced joints, followed by finite element-based simulations on the validated girders. Extensive finite element analyses were conducted to investigate the influence of the width and thickness parameters of the splice plates under both three-point and four-point loading. Under both these loading conditions, superior performance was attained when a thicker flange splice and a thinner web splice were adopted. Strength improvements of 10% and 4% were achieved when the width of the flange splice (WFS) was dropped from 140 mm to 50 mm while maintaining a constant width of the web splice (WWS) at 270 mm, under three-point and four-point loading conditions respectively. Capacity enhancements of 16% and 23% were noted when WWS was raised from 60 mm to 270 mm while maintaining a constant WFS at 140 mm, under the above mentioned loading conditions, respectively. Lastly, the feasibility of artificial neural networks (ANNs) to predict the ultimate capacity of the splice joints was assessed, and it was verified that they could predict the ultimate load capacity with reasonable accuracy. The predicted load capacity of the splice joint outside the training set revealed that an appropriately trained and optimized ANN network could reliably estimate the strengths with a mean absolute error of less than 2%.
... This ultimately indicates that machine learning models have significant potential for accurate prediction of the strength parameter of concrete. Further, these studies showed that machine learning models are a reliable tool to predict other characteristics of concrete such as flexural strength (Behnood and Golafshani 2020), split tensile strength (Behnood et al. 2015), shear strength of beams (Naderpour et al. 2018), slump (Azimi-Pour et al. 2020, and modulus of elasticity (Tavana Amlashi et al. 2021). ...
Compressive strength (CS) is a key property for the practical application of HPC but the determination of CS by traditional experimentation is laborious and time consuming. In order to precisely predict the compressive strength, machine learning (ML) techniques were employed, and the development of a large database was a crucial prerequisite. Gaussian process Regression (GPR), Decision Tree (DT), Multi-linear Regression (MLR), Support Vector Machine (SVM), and Bagged Regression (BR) were used for compressive strength prediction with an R² value of 0.72–0.94. The comparison of MLR, DT, SVM, BR, and GPR shows that GPR model for prediction of compressive strength has the highest performance (R = 0.971, R² = 0.943) and the least prediction error (RMSE = 4.397, MAE = 3.230). The best model interpretation (GPR model) indicates that cement, coarse aggregate, sand, and water were the most influential features. Furthermore, concrete ingredients showed positive (e.g., cement, sand, age) and negative (e.g., water, superplasticizer) relationships with compressive strength. The model accuracy showed the best performance when compared with different literature-reported models. Thus, this study provides a sustainable and cost-effective approach to estimating the compressive strength of HPC.
... Where ∑ N r=1 w rj is the sum of the connection weights between the N input neurons and the hidden neuron j, and ϑ jk is the connection weight between the hidden neuron j and the output neuron k [37][38][39]. Fig. 5 illustrates the plot of the sensitivity analysis results. As shown, the influence of most of the parameters on the damage level is similar. ...
Fire poses a major risk to the structural integrity of a bridge or progressively to the bridge's failure. This study uses an Artificial Neural Network (ANN) to investigate the determining factors in bridge fire incidents. Implementation of this powerful model produces an estimation of the damage levels. The basis of the proposed model is determined by critical factors, including bridge location, materials, structural systems, annual average daily traffic (AADT), ignition source, combustible type, and bridge face exposed to fire. Results show that steel I-girder bridges are the most susceptible structural system in a fire. Moreover, a fire involving tankers containing hydrocarbon fuel and trucks with solid combustible cargo is the most dangerous to a bridge. The accuracy of the proposed model is verified, and its outcome can be utilized to determine fire risk based on discrete characteristics. Based on the proposed model, a combination of hazard prevention and protection measures may be utilized to improve structural integrity, human life safety, and durability, and to reduce maintenance costs.
... Some studies have evaluated the shear strength of FRPconcrete beams using different ML models. Naderpour et al. (2018) used ANN to predict the shear strength of concrete beams reinforced with FRP bars. They compiled 110 tested data sets of FRP-concrete beams to train the ANN model with obtained R 2 value of 0.93 and RMSE of 11.36 kN. ...
Shear strength is a very important parameter in designing of reinforced concrete beams or concrete beams reinforced with fiber-reinforced polymer (FRP) bars. So far, numerous studies and design codes have proposed empirical-based formulas for predicting the shear strength of FRP-concrete beams. However, a difference exists between the proposed formulas and experimental results. This study predicts the shear strength of FRP-concrete beams using the novel hybrid BR-ANN model, which integrates artificial neural network (ANN) and Bayesian regularization (BR). For that, a comprehensive database consisting of 303 experimental results is compiled for developing the BR-ANN models. The performance results of BR-ANN are compared with those of 15 existing empirical formulas, which were proposed in typical design codes and well-known published studies. The predicted outputs are evaluated utilizing indicators, which are goodness of fit (), root mean squared error (), and mean value of the ratio . The results reveal that the BR-ANN model outperforms other empirical formulas with a very high (0.987), very small (7.3 kN). In addition, the mean value of the ratio is equal to unity. Moreover, effects of input variables on the shear strength are evaluated. Finally, a practical design tool is developed to apply the BR-ANN model in calculating the shear strength of FRP-concrete beams.
... Existing empirical or semi-empirical models for shear strength prediction of FRP-RC members have been established by simply modifying the methods available for steel-RC members and based on limited test data [5][6][7][8][9]. This causes a wide scatter in the prediction when compared to experimental data, and it tends to be conservative, demanding a large amount of FRP and making FRP applications less costeffective [10][11][12]. Nevertheless, there has been a recent tendency among researchers to enhance the amount of data used in their study, and numerous recent studies have used machine learning (ML) approaches to estimate the shear resistance of slender and deep reinforced concrete beams. ...
The shear strength prediction of a fiber-reinforced polymer (FRP)-reinforced concrete (RC) beam is a difficult undertaking that is influenced by several design variables. While the usage of FRP bars has emerged as a viable alternative in reducing corrosion problems associated with steel reinforcement in an extreme environment, a precise and reliable method of shear strength prediction is required to assure cost-effective material use and optimum design. Several design provisions and optimized design equations are now available in the current literature; nevertheless, when these equations are used, a significant variation between the experimental and the predicted shear strength of FRP-RC beams is observed. This study utilised the power of data-driven modelling techniques for enhanced prediction capability of such a complex phenomenon. The objective of the current study is to develop a data-driven shear strength prediction model for FRP-RC slender beams using the extreme gradient boosting (XGBoost) algorithm. A large database of 302 tests of RC beams longitudinally reinforced with FRP bars without stirrups was collected from the available literature. The performance of the proposed ML model is compared with the existing standards, codes, guidelines, and optimized shear strength equation. The results reveal that the XGBoost model outperformed the existing shear strength provisions and has a high level of prediction accuracy.KeywordsFRP-reinforced concrete beamsShear strengthMachine learning
... Compressive and flexural strength of SFRC KNN, LNR, SVM, ANN, boosting [164] 2021 Post-cracking tensile strength of fiber-reinforced concrete ANN [165] 2021 Ultimate strength of FRP-confined concrete SVM [166] 2021 Compressive and tensile strength of HPC SVM, ANN, boosting [80] 2021 Compressive strength of recycled aggregate concrete ANN, SVM [167] 2022 Mechanical properties of composite laminate Shear strength of SFRC beams ANN [169] 2010 Compressive strength of FRP-confined concrete columns ANN [170] 2012 Buckling and post-buckling loads of compression members ANN [171] 2013 Failure mode, shear strength and deformation capacity of infilled walls ANN [172] 2014 Compressive strength and strain of FRP-confined columns ANN [173] 2014 Shear strength of FRP-reinforced concrete flexural members without stirrups ANN [174] 2014 Shear strength of RC beam-column joints LNR, symbolic regression [175] 2016 Punching shear capacity of FRP-reinforced concrete slabs SVM [176] 2017 Compressive strength of FRP-confined concrete circular columns ANN[84] 2018 Shear resistance of FRP bars-reinforced concrete beams ANN[85] 2018 Failure mode and shear strength of RC beam-column joints LR, LNR, KNN, Naïve Bayes, SVM, DT, bagging[177] 2018 Shear strength of SFRC beams SVM[178] 2018 Shear strength of squat RC shear walls ANN[179] 2019 Punching shear capacity of SFRC slabs LNR, ANN[86] 2019 Load-carrying capacity and mode failure of beam-column joint Extreme learning machine[180] 2019 Failure mode of circular RC bridge columns KNN, DT, Naïve Bayes, ANN, bagging[87] 2019 Thermal and structural response of RC members ANN[181] 2019 Axial compression capacity of SCFST columns ANN[88] 2019 Shear strength of steel fiber-unconfined RC beams SVM[105] 2019 Shear strength of SFRC beams multi-expression programming[182] 2019 Failure modes of ductile and non-ductile concrete joints DT[99] 2019 In-plane failure modes of infilled RC walls DT, LR, ANN, RF, SVM, boosting[183] 2020 Seismic failure mode identification of RC shear walls Naïve Bayes, KNN, DT, bagging, boosting[184] 2020 Failure mode and bearing capacity of RC columns boosting[89] 2020 Shear capacity of one-way slabs under concentrated loads ANN[185] 2020 Failure mode and shear capacity of UHPC beams SVM, ANN, genetic programming, KNN[90] 2020 Shear capacity of deep RC beams SVM, ANN, boosting[186] 2020 Shear strength of internal RC beam-column joints boosting[187] 2020 Shear strength of SFRCB without stirrups SVM[188] 2020 Axial compression capacity of circular CFST with UHPC ANN[91] 2021 Plastic hinge length of RC columns boosting[189] 2021 Punching shear strength of flat RC slabs without transverse reinforcement LNR, SVM, DT, KNN, bagging, boosting (continued on next page) ...
With the informatization of the building and infrastructure industry, conventional analysis methods are gradually proving inadequate in meeting the demands of the new era, such as intelligent synchronization and real-time simulation. Artificial intelligence (AI) technology has emerged as a promising alternative due to its high expressiveness, efficiency, and scalability. This has given rise to a new research field of AI-based computation in civil engineering. In this study, a state-of-the-art review of the research on material and structural analyses using AI technology in civil engineering was performed to provide a general introduction to the current progress. The research was classified into static feature studies, dynamic feature studies, and composite feature studies according to the problem inputs. The general methodology, commonly used AI models, and representative applications of each research category were elaborated. On these bases, the strengths and weaknesses of current studies were discussed. To demonstrate the accuracy and efficiency of AI models in comparison with conventional numerical methods, a concrete example of an end-to-end deep learning framework for structural analysis was highlighted. Finally, we suggested four open problems from the perspective of engineering applications, indicating the major challenges and future research directions regarding AI-based computational analysis in civil engineering.
... For both the number of hidden layers and their associated nodes, there is no reliable principle. Some studies (Kaveh & Khalegi, 2000;Naderpour, 2018;Naderpour et al., 2018) showed that a neural network with one hidden layer can be sufficient for achieving good results, so the proposed ANN model also included one hidden layer. In this study, the structure of the proposed ANN model comprises the following parameters: The number of input layer neurons is six, the number of hidden layer neurons is seven, and one neuron from the output layer is used as shown in Fig. 1. ...
A practical artificial neural network tool is proposed for predicting the ultimate bending moments of reinforced concrete beams strengthened by the techniques of externally bonded fiber-reinforced polymer and near-surface mounted fiber-reinforced polymer. Accordingly, the testing database of 131 specimens was gathered for use in developing the artificial neural network model. In this regard, the breadth and height of the beam section, the compression strength of the concrete, the ratio of material reinforcement, and the elastic modulus of fiber reinforced polymer were regarded as input variables, whereas the ultimate bending moment was regarded as an output variable. The performance of the proposed artificial neural network model was compared to the current design model of the American Concrete Institute guide. The comparative analysis demonstrated that the proposed model made more accurate predictions than the current model. Based on the proposed model, a graphical user interface was created to facilitate the prediction of the ultimate bending moments of reinforced concrete beams with fiber-reinforced polymer strengthening.
... Furthermore, exposing structures (e.g., water treatment facilities, marine structures, and bridges) to extreme corrosion compromises their structure and, therefore, reduces their service life tremendously [2]. Fiberreinforced polymer (FRP) bars are a promising substitute for traditional reinforcing steel bars [3,4]. Additionally, the greater strength, smaller weight, and higher axial stifness-to-weight ratio of FRPs make them more attractive solutions. ...
The shear strength of fiber-reinforced polymer (FRP) reinforced concrete beams is often given a large safety margin by current construction requirements. Six characteristics are utilized as inputs to compute the shear strength of FRP-reinforced concrete beams. This study uses 198 samples from the literature to predict the shear strength of 139 training samples and 59 testing samples. Additionally, the ANN structure is optimized with the firefly algorithm. The FA-ANN model is also compared to ACI-440, CSA-S806, and BISE-99 codes, and the optimized model by Nehdi et al. Findings show that regarding the shear strength of FRP-reinforced concrete beams, the firefly algorithm-optimized model performs better than the other four models. Concerning accuracy, the coefficient of correlation, R2, was calculated as 0.961, while the average absolute error (AAE) is 0.22 for the shear strength of FRP-reinforced beams.
... The results show that, compared with ordinary concrete, UHPC with steel fibers can significantly improve the cracking stiffness and bending resistance of the beam. To study the shear resistance of UHPC beams, many studies have proposed empirical formulas to estimate the shear strength of UHPC beams [23][24][25][26] , but the experimental results and the results of the prediction formula often differ 27 . These empirical formulas are aimed at specific experimental research. ...
Currently, concrete structures have increasingly higher requirements for the shear capacity of beams, and ultrahigh-performance concrete (UHPC) beams are increasingly widely used. To facilitate the design of UHPC beams, this paper constructs a UHPC beam shear strength prediction model. First, static shear tests were conducted on 6 UHPC beam specimens with a length of 2 m and a cross-sectional size of 200 mm × 300 mm to explore the effects of the UHPC strength, shear span ratio, hoop ratio, and steel fiber content on the shear resistance and failure morphology of the UHPC beams. Based on the results of this study and a static load experiment of 102 UHPC beams in the literature, the construction includes the shear span ratio (λ), beam section width (b), beam section height (h), hoop ratio (ρSV), UHPC compressive strength (fc), steel fiber volume fraction (Vf), and the UHPC beam shear capacity (Vex) 7 parameter database. Based on the construction of the database, 1200 BPNN models were trained through trial and error. The models were evaluated using the correlation coefficient R, root mean square error RMSE, and a20-index indicators, and the optimal BPNN model (6-15-8-1) was determined based on the ranking of RMSE. After the optimal BPNN is optimized by a genetic algorithm, the prediction performance of the model is improved. The correlation coefficient between the predicted value and the experimental value is R² = 0.98667, and RMSE = 7.38. This model can reliably predict the shear strength of UHPC beams and provide designers with a reference for the design of UHPC beams. Finally, after sensitivity analysis, the influence of each input parameter on the UHPC shear capacity is determined.
... The conventional procedure of lab-scale testing provides a good and reliable estimate of compressive strength, but these methods require a large amount of time, materials, and resources, which makes their practical employability difficult. The recent development in machine learning (ML) provides good and reliable tools to accurately predict the properties of concrete such as slump [4], shear strength of beams [5], bond strength [6], fire resistance of concrete columns [7], flexural strength [8], split tensile strength [9], modulus of elasticity [10], etc. ...
Compressive strength determination of high-performance concrete (HPC) is necessary for its practical usage. However, experimental testing for this purpose is resource intensive and time-consuming. Recently, machine learning has emerged in this field, especially data-driven modeling. In this study, a novel hybrid model for the compressive strength prediction of HPC is developed using the cascade forward neural network (CFNN) and artificial bee colony (ABC) optimization. The hybrid model used the ABC optimization method to select the optimal architecture of the neural network. A comprehensive database of 2171 data points containing information about cement, blast furnace slag, fly ash, water, coarse aggregate, sand, and age as input variables, and compressive strength as output variable is used to develop the model. Results indicated that the optimal neural network architecture selected by the ABC method consists of 2 layers and the developed model (CFNN-ABC) could accurately predict the compressive strength of HPC with correlation (R) and determination coefficients (R2) of 0.976 and 0.953, respectively. The feature importance of the model revealed that the cement and sand were more influential features as compared to the other features. The partial dependence analysis demonstrated the effect of variation in input parameters on the attained compressive strength. Furthermore, the model validation with previously developed models using performance indices showed that the proposed hybrid model outperformed other models in all performance indices including root mean square error (RMSE ~ 4.04) and mean absolute error (MAE ~ 3.10). Therefore, the present work provides a novel and efficient option to predict the compressive strength of HPC which can aid in the design of sustainable infrastructures without going through costly and time-intensive experimentation.
... Artificial neural networks (ANNs) have seen extensive structural engineering applications for modeling and approximating complex nonlinear and multi-dimensional relationships, with the benefit of not requiring knowledge of fundamental mechanics or the nature of the relationship. Such as approximation of structural responses [23][24][25][26][27][28]; system identification [29,30]; engineering design [31][32][33]; structural health monitoring [34][35][36][37][38], etc. ANNs can be trained with sample cases of input-output datasets (i.e., patterns or instances) to predict output for similar other input datasets that networks never encountered in training. Fig. 8 shows the multi-layer feedforward network architecture optimized in this study as the proposed prediction model for the patchloading resistance of CWGs. ...
Corrugated-web steel I-girders (CWGs) have better shear and torsional strengths, offering additional stability against out-of-plane web buckling and local crippling under patch loads than flat web girders. CWGs are lightweight and more cost effective than flat web beams for applications in buildings and bridges. The existing design models for estimating patch-loading resistance in CWGs assume equal widths in web corrugation folds for a simply supported span. This study examined the parametric influence of unequal trapezoidal-web folds on the nonlinear behavior of CWGs under compressive patch loading. A total of 533 steel I-girders with simply supported (SS) and 478 cantilever spans (CS) were analyzed using a calibrated nonlinear finite element (FE) methodology. The generated datasets of CWGs consisting of governing parameters (i.e., girder geometry and material properties) were mapped to refined patch-loading resistance using artificial neural networks (ANNs) for both CS and SS-type spans. The proposed ANN formulation showed better prediction accuracy (mean absolute error of about 2–4% in this study) than the existing design models for estimating the patch-loading resistance of CWGs. For practical applications and conservative patch-loading resistance, prediction modification factors were also proposed for ANN formulation and existing design models within a targeted error margin.
... In another study, Duan et al. [155] evaluated the com- (continued on next page) (continued on next page) (continued on next page) pressive strength of concrete containing recycled aggregate (coarse aggregate) by 14 different inputs using the BPNN method and concluded that this method could be a suitable tool for predicting RAC compressive strength. Also, Naderpour et al. [156] evaluated the compressive strength of concrete containing construction waste (coarse-grained alternative) by the BPNN method (including 6 inputs and 18 hidden nodes) and evaluated the effect of each input parameter on the compressive strength. Their results indicate that water absorption as one of the inputs of the BPNN model has the most significant effect on compressive strength compared to other inputs. ...
Concrete is one of the most widely used materials in various civil engineering applications. Its global production rate is increasing to meet demand. Mechanical properties of concrete are among important parameters in designing and evaluating its performance. Over the past few decades, machine learning has been used to model real-world problems. Machine learning, as a branch of artificial intelligence, is gaining popularity in many scientific fields such as robotics, statistics, bioinformatics, computer science, and construction materials. Machine learning has many advantages over statistical and experimental models, such as optimal accuracy, high-performance speed, responsiveness in complex environments, and economic cost-effectiveness. Recently, more researchers are looking into deep learning, which is a group of machine learning algorithms, as a powerful method in matters of diagnosis and classification. Hence, this paper provides a review of successful ML and DL model applications to predict concrete mechanical properties. Several modeling algorithms were reviewed highlighting their applications, performance, current knowledge gaps, and suggestions for future research. This paper will assist construction material engineers and researchers in selecting suitable and accurate techniques that fit their applications.
In mountainous regions, bamboo has gained popularity as a local reinforcement material for concrete due to its availability and cost-effectiveness. However, the lack of standardized design guidelines complicates the assessment of deflection behavior under varying loads. Challenges such as bond failure, shrinkage, corrosion, and material strength uncertainties further limit its structural application. To address these issues, this study employs an extreme learning machine to predict deflection in plain cement concrete beams, steel-reinforced cement concrete beams, and bamboo-reinforced concrete beams. The performance of extreme learning machine is compared with support vector regression and artificial neural networks. A total of 122 specimens were analyzed using input features such as cement content, water-cement ratio, fine aggregate, coarse aggregate, tensile strength, concrete compressive strength, percentage of reinforcement, applied load, cross-sectional area, and clear cover. The extreme learning machine method demonstrated satisfactory performance with a coefficient of deformation of 0.9975, Mean Square Error of 0.0003, Mean Absolute Error of 0.0052, Mean Absolute Percentage Error of 0.0998, and Root Mean Square Error of 0.01694. These results suggest that extreme learning machine is an efficient machine learning algorithm for predicting the deflection behavior of different reinforced concrete beams, providing a reliable alternative to traditional methods.
Purpose
The application of steel fiber reinforced concrete (SFRC) beams is limited in practice, partially due to the lack of accurate shear strength prediction models. This study aims to develop a reliable shear strength prediction model for SFRC beams.
Design/methodology/approach
In this study, an artificial neural network was employed to predict the shear strength of SFRC beams, utilizing a comprehensive database of 562 experimental studies. Multiple neural networks were established with varying hyperparameters, and their performance was evaluated using statistical parameters.
Findings
The neural network with 11 neurons showed superior results than other networks. The performance evaluation, efficiency and accuracy of the selected neural network were examined using margin of deviation, k-fold cross-validation, Shapley analysis, sensitivity analysis and parametric analysis. The proposed artificial neural network model accurately predicts the shear strength and outperforms other existing equations.
Originality/value
This research contributes to overcoming the limitations of existing prediction models for shear strength of SFRC beams without stirrups by developing a highly accurate model based on ANN. Utilizing a comprehensive database and rigorous evaluation techniques enhances the reliability and applicability of the proposed model in practical engineering applications.
To enhance the durability of concrete structures in harsh environments, replacing steel bars with FRP bars is judged a feasible method. The low elastic modulus and transverse strength of FRP bars result in the weak shear capacity of concrete beams reinforced with FRP bars. Reviewing and summarizing existing literature are important for addressing this challenge. The research progress on the shear behaviour of concrete beams reinforced with FRP bars was systematically described from two aspects in this paper. In the aspect of literature review, shear mechanisms and main influencing factors of concrete and FRP stirrups were summarized. In addition, achievements and shortcomings of existing literature were summarized, and potential research directions were pointed out. In the aspect of design method evaluation, a database containing 525 samples was established and calculation models of shear capacity of concrete beams reinforced with FRP bars in seven codes were evaluated. Evaluation parameters mainly included shear span ratio, effective height, and maximum strain of FRP stirrups. Based on the calculation model in Italian code CNR DT203-06, an optimized model was proposed to improve the accuracy in predicting concrete shear contribution. The review and evaluation in this paper had important reference significance for improving the design level of concrete structures reinforced with FRP bars and promoting the engineering application of FRP bars.
It is widely recognized that critical crack angle θ is a pre-requisite to calculate the shear capacity of RC elements in traditional modified compression field theory (MCFT), and it is often determined by an iterative calculation or by presupposing an empirical value. This study proposes a straightforward solution of critical crack angle aided by machine learning. Firstly, 215 reinforced concrete T-girder test samples are collected from published literatures, and are analyzed by traditional MCFT to determine their shear capacity and their corresponding critical compressive strain of the upper edge concrete εtom and the critical crack angle θ. Subsequently, integrated BP (back propagation) neural network models are established to seek for a quantitative regression between the iteratively obtained θ, εtom and other MCFT input parameters. Finally, the obtained regression equations are incorporated into traditional MCFT framework to determine the shear capacity straightforwardly. The results indicate that critical crack angle θ of reinforced concrete T-girder exponentially grows by increasing the strength eigenvalue ρv·fyv/fc or decreasing the longitudinal reinforcement ratio ρl. While the compressive strain of concrete in the compression region εtom exhibits a logarithmic function with the strength eigenvalue ρv·fyv/fc and the shear span ratio λ. The proposed straightforward calculation approach is superior to other methods both in efficiency and accuracy. Specifically, the goodness-of-fit of the proposed approach is 1.7-fold higher than that of the American ACI318-14, and the coefficient of variation is reduced by 43% compared to the European EN 1992-1-1;2004.
The enhanced mechanical and durability properties of Engineered Cementitious composite (ECC) make it famous worldwide. However, due to the unavailability of material design guides, its mixed design is still based on extensive experimentation, which is a highly uneconomical and time-consuming process. This study aims to develop a machine learning-based model capable of predicting the suitable mix design for ECC. Initially, a dataset of 147 data points composed of mixed design and their associated stress-strain curves was collected from the published literature. This paper uses an Artificial Neural Network (ANN) model incorporating 10 input parameters, including all the ingredients of the mix design and the properties of the fibers utilized to predict the complete tensile stress-strain curve of ECC in terms of the bilinear idealized curve. Furthermore, a detailed sensitivity analysis was also conducted to study the influence of each parameter on the tensile behavior of ECC. This model gives RMSE for a training set of yield stress, yield strain, ultimate stress, and ultimate strain as 0.042, 0.002, 0.36, and 0.3, respectively, while for the testing set, RMSE is 0.2, 0.004, 0.374, 0.3133 and 0.18, respectively. Finally, the model's performance was revalidated experimentally by employing mixes that were not part of previous data. The results show that this model possesses a high accuracy and can help to design ECC without extensive experimentation, which can help in the advancement of the commercialization potential of this robust composite.
Purpose: In recent years, stainless steel reinforcement has gained popularity in the construction industry due to its unique properties such as excellent durability, long life cycle, and low maintenance and monitoring requirements. Furthermore, stainless steel possesses exceptional corrosion resistance and remarkable mechanical properties such as significant strain hardening and high ductility. Bond, a crucial parameter in the design of reinforced concrete members, especially those containing stainless steel reinforcement, needs to be accurately assessed. Study design/methodology/approach: This research aims to utilize Artificial Neural Networks (ANN) to predict the ultimate bond strength between concrete and stainless steel reinforcement. To train, validate, and test the ANN models, 101 pull-out test samples were used. The data includes a wide range of key influential parameters, such as different stainless steel grades, concrete strengths, and geometric properties. The ANN model is calibrated and validated using existing data available in the literature. Findings: The model's predictions for the training, validation, and testing data exhibit a remarkable level of agreement and fit with the actual data with regression values of 0.9786, 0.9754 and 0.9546 and RMSE values of 0.11, 0.12 and 0.15, respectively. Originality/value: The current design standards such as Eurocode 2 do not include specific bond design rules for stainless steel reinforcement. The rapid advancement of Artificial neural networks has provided an efficient and reliable approach for predicting various structural performance of structural members, through the development of design formulas that can account for various influential parameters. The results and observations indicate that the proposed model provides an accurate and efficient predictions of bond strength for stainless steel reinforcements.
The use of steel shear walls is a common practice in strengthening buildings against lateral loads. Recent studies have explored the use of corrugated sheets as alternatives to smooth sheets with stiffeners. Architectural and structural requirements often necessitate openings in shear wall sheets, which, when incorporated, reduce the structural resistance of the steel shear wall due to their size and position. Calculating the extent of this reduced resistance through rigorous nonlinear practical analysis is time-consuming. There is a need for a rapid and accurate method.
In this research, initially, utilizing 130 collected data points and finite element analysis of various-sized and positioned openings, a neural network is trained to estimate the percentage reduction in shear wall resistance. Subsequently, employing 400 extracted data points from the trained network, a relationship is proposed to estimate the magnitude of reduction in shear wall resistance.
Composites have a wide range of applications across various industries due to their high strength-to-weight ratio, corrosion resistance, durability, versatility, and lightweight structures. However, manufacturing reinforced composites and the various tests they undergo for their appropriate applications are extensive and expensive. Because of this, many researchers have employed the machine learning (ML) technique to evaluate the significance of the process parameters and predict the properties for effective composite design and their widespread applications. Therefore, this study critically reviewed and compared the different ML models applied to predict the mechanical, thermal, tribological, acoustic, and electrical properties of different reinforced composites. ML models, their appropriate methods, database size and source, training and testing data, input and output parameters, and statistical index are also summarized. In addition, the performance evaluation of ML models and statistical indexes of different property predictions is critically analyzed based on several models' training and testing scores, which may help select appropriate ML models to predict reinforced composite properties. This review study can provide insights into applying ML techniques to reinforced composites to develop novel, innovative composites.
Advanced concrete technology is the science of efficient, cost‐effective, and safe design in civil engineering projects. Engineers and concrete designers are generally faced with the slightest change in the conditions or objectives of the project, which makes it challenging to choose the optimal design among several ones. Besides, the experimental examination of all of them requires time and high costs. Hence, an efficient approach is to utilize artificial intelligence (AI) techniques to predict and optimize real‐world problems in concrete technology. Despite the large body of publications in this field, there are few comprehensive surveys that conduct scientometric analysis. This paper provides a state‐of‐the‐art review that lists, summarizes, and categorizes the most widely used machine learning methods, meta‐heuristic algorithms, and hybrid approaches to concrete issues. To this end, 457 publications are considered during the recent decade with a scientometric approach to highlight the annual trend/active journals/top researchers/co‐occurrence of key title words/countries' participation/research hotspots. In addition, AI techniques are classified into distinct clusters using VOSviewer clustering visualization to identify the application scope and their relationship through the link strength. The findings can be a beacon to help researchers utilize AI techniques in future research on advanced concrete technology.
Making physical sense of the shear behavior of reinforced concrete (RC) elements remains an open challenge and an ongoing investigation that often leads to upgrades of design codes/provisions. The physically sound critical shear crack theory (CSCT) which forms the basis for shear design in some popular design codes has been mainly successful in steel-RC members, but not adapted to concrete members reinforced with fiber reinforced polymer (FRP) bars, given that the shear behavior of the latter is significantly less comprehended compared to that of the former. The present study, therefore, aims to develop a model for the shear strength of FRP-RC elements by refining the CSCT parameters. An extensive experimental database is compiled with a total of 420 shear-controlled slender FRP-RC beams without stirrups from 56 investigations – the largest database of this structural material type ever compiled. In addition to being physically sound, the strength predictions by the developed model are found to be accurate, compared to existing design methods. It is adaptable for predicting the shear strength of slender FRP-RC as well as being valid for steel-RC elements. The new model’s size effect consistency is verified, and it proves to be more robust than that of the traditional CSCT. In the case of FRP-RC members, the latter wrongly signifies size effect behavior that mostly obeys the linear elastic fracture mechanics (which is believed to be too strong for concrete) while the former predicts a more physically logical behavior similar to that of the nonlinear fracture mechanics.
Retrofitting of reinforced concrete beams using fiber-reinforced polymer (FRP) sheets is one of the most commonly used methods to enhance shear and bending behavior. A comprehensive review shows that the presented formulas are insufficiently accurate to predict the shear contribution of FRP sheets. Researchers have also studied only 45° and 90° for the FRP sheets in previous research. This paper presents a comprehensive formula for evaluating the shear contribution of full- and U-wrap FRP sheets. For this purpose, experimental data are collected for angles of 45° and 90°. Then, two experimental tests are modeled using ABAQUS software. After verifying the models, the shear contribution of FRP sheets is determined for different values of the input parameters and at the angles of 30°, 60° and 75°. Therefore, to present a formula for evaluating the shear contribution of FRP sheets, multigene genetic programming is used. The results show that the presented formula has sufficient accuracy in computing the shear contribution of FRP sheets at angles of 30°, 45°, 60°, 75° and 90° in comparison with other existing formulas from codes and literature. The effect of width, thickness, the ultimate strain of FRP, effective depth of FRP, beam shear span length, the effective depth of the concrete beam, spacing between the FRP strips, compressive strengths, angle of FRP sheets, and elasticity modulus of FRP sheets is considered on computing the shear contribution of FRP sheets. The results show that the mean absolute percentage error (MAPE), root mean square error (RMSE), and R2 for the proposed formulas are 12.48, 9.84 and 0.98, respectively, which performs better than the other existing methods.
The corrosion of reinforced concrete (RC) structures is of increasing concern across the world. Repair, restoration, replacement, and the construction of new buildings all necessitate the use of economic and long‐lasting technology. Fiber reinforced polymer (FRP) has been widely used in both retrofitting and new construction of buildings. FRP has seen an increase in the application as a repair composite material in reinforced concrete and masonry structures over the last decade due to its many properties. This material has various benefits, including excellent strength‐to‐weight ratios, stiffness‐to‐weight, lightweight, potential long‐term durability, and relative ease of field use. This chapter provides a summary of the machine learning models in the estimation of bond strength between FRP and concrete surface, shear, and flexural strength of FRP‐wrapped reinforced concrete beams.
Performance of the existing design provisions on fiber reinforced polymer (FRP) reinforced concrete (RC) – FRP-RC – deep beams without stirrups appear to be less robust than when applied to slender geometries. This, coupled with the need for making physical sense of the shear behavior of FRP-RC elements, motivates the present work to extend the Critical Shear Crack Theory to the former, given that the original theory forms the basis for the Swiss design code and the new draft of the Eurocode for shear and punching shear designs. Attributes of deep members such as the geometry of the critical shear crack, the critical section location and the arch action are incorporated into the modified model using special functions and parameters obtained via nonlinear regression, resulting in a model justified with more mechanics and less empiricism than most of the existing design equations. A comprehensive database of deep FRP-RC beams without stirrups was compiled and its behavior thoroughly critiqued with respect to various effective shear strength parameters. The database artifacts and their connection to the less satisfactory performance of the codes/guidelines are discussed. Performance of the developed model was successfully validated using both the training data and blind simulation against the database.
The computer aided models have received much attention in the recent years for solving diverse civil engineering applications. In the current review, the applications of artificial intelligence (AI) methodologies in modeling beam shear strength are presented. The review is attempted to give an insightful version for AI models progression in modeling diverse shear strength of structure beams types. Several essential components are surveyed and summarized including, the types of the applied AI models, the input parameters used for the development, types of the additive material, number of the modelled dataset, performance metrics, sources of the modelled data and data availability. Based on the reported literature, several significant core aspects on the AI model’s implementation are assessed and discussed scientifically. Future research directions are presented with respect to the exhibited literature review.
Recent years have witnessed a surge in the application of machine learning techniques for solving hard to solve structural engineering problems. The application of machine learning can replace the use of empirical and semi-empirical prediction models currently used in practice with highly accurate models. This paper provides a detailed discussion on the basic terminologies and concepts of commonly used machine learning algorithms for solving structural engineering problems. To provide confidence to use this method and show the potential of machine learning in accurately predicting the results of complex civil engineering problems, a comprehensive literature review on the application of machine learning in shear strength prediction is also presented. The literature review covers the application of different machine learning algorithms in predicting the shear strength of conventional concrete beams, steel fibre reinforced concrete beams, beams reinforced with FRP bars as well as high strength concrete beams. Major observations, challenges and future scope in this field are also discussed in detail. This article will be a valuable resource for individuals who are unfamiliar with machine learning yet aspire to learn more about it.
In reinforced concrete structures, the utilization of composite rebar has been increased by considering their high corrosion resistance, anti-magnetic properties, and significant tensile strength. According to the lower elasticity modulus of composite rebar in comparison with steel rebar, concrete beams reinforced including composite rebar possess a relatively lower shear strength by comparing with steel rebar. In addition, in concrete beam, reinforced shear failure by composite rebar is commonly brittle and requires precise performance prediction of the members. Thus, the reinforced concrete beams' shear strength by composite rebar is predicted utilizing an Extreme Learning Machine network based on Chaos Red Fox Optimization Algorithm (ELM-CRFOA) according to a wide range of data. The most important parameters, which are considered in this investigation, are the web width, beam effective depth, the strength of concrete compressive, the ratio of the shear span to depth, FRP longitudinal bars elasticity modulus, and ratio of the longitudinal reinforcement. This method's precision has been proved by having a comparison among the model predictions and the accumulated data and available shear design equations. According to the study outcomes, the presented model has precise outcomes in computing the concrete beams' shear strength in comparison with other existing relations. For assessing input parameters' impact on the FRP-reinforced concrete beams' shear strength, a sensitivity analysis is executed.
Previous studies on machine-learning (ML) prediction of axial capacity of composite concrete-filled steel tubular (CFST) columns under axial loading relate mainly to only one cross-section shape, meaning that they are limited to only one given application. In this paper, a ML model—namely support vector machine (SVM)—is proposed for the prediction of CFST columns with different cross-section shapes: circular, elliptical, square and rectangular, because they are the most widely used in engineering structures. A database consisting of 1093 tests was gathered from the available literature and used to train and validate the SVM model. The model’s performance was quantified by various performance indicators: coefficient of determination (R2), root mean squared error, mean absolute error, Willmott’s index of agreement, and mean absolute percentage error. Based on the SVM model, sensitivity analysis, influence of different factors, parametric study and comparison with the literature are presented. A graphic user interface of the proposed model was also implemented. The model could be extended to study the effect of other cross-sectional shapes, making for wider applicability.
Corrosion of steel reinforcement is considered as the main cause of concrete structures deterioration, especially those under humid environmental conditions. Hence, fiber reinforced polymer (FRP) bars are being increasingly used as a replacement for conventional steel owing to their non-corrodible characteristics. However, predicting the shear strength of beams reinforced with FRP bars still challenging due to the lack of robust shear theory. Thus, this paper aims to develop an explicit data driven based model to predict the shear strength of FRP reinforced beams using multi-objective evolutionary polynomial regression analysis (MOGA-EPR) as data driven models learn the behavior from the input data without the need to employee a theory that aid the derivation, and thus they have an enhanced accuracy. This study also evaluates the accuracy of predictive models of shear strength of FRP reinforced concrete beams employed by different design codes by calculating and comparing the values of the mean absolute error (MAE), root mean square error (RMSE), mean (u), standard deviation of the mean (o), coefficient of determination (R2), and percentage of prediction within error range of +-20% (a20-index). Experimental database has been developed and employed in the model learning, validation, and accuracy examination. The statistical analysis illustrated the robustness of the developed model with MAE, RMSE, u, o, R2, and a20-index of 14.6, 20.8, 1.05, 0.27, 0.85, and 0.61, respectively for training data and 10.4, 14.1, 0.98, 0.25, 0.94, and 0.60, respectively for validation data. Furthermore, the developed model achieved much better predictions than the standard predictive models as it scored lower MAE, RMSE, and o, and higher R2 and a20-index. The new model can be used in future with confidence in optimized designs as its accuracy is higher than standard predictive models.
Compressive strength of concrete, recognized as one of the most significant mechanical properties of concrete, is identified as one of the most essential factors for the quality assurance of concrete. In the current study, three different data-driven models, i.e., Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Multiple Linear Regression (MLR) were used to predict the 28 days compressive strength of recycled aggregate concrete (RAC). Recycled aggregate is the current need of the hour owing to its environmental pleasant aspect of re-using the wastes due to construction. 14 different input parameters, including both dimensional and non-dimensional parameters, were used in this study for predicting the 28 days compressive strength of concrete. The present study concluded that estimation of 28 days compressive strength of recycled aggregate concrete was performed better by ANN and ANFIS in comparison to MLR. In other words, comparing the test step of all the three models, it can be concluded that the MLR model is better to be utilized for preliminary mix design of concrete, and ANN and ANFIS models are suggested to be used in the mix design optimization and in the case of higher accuracy necessities. In addition, the performance of data-driven models with and without the non-dimensional parameters is explored. It was observed that the data-driven models show better accuracy when the non-dimensional parameters were used as additional input parameters. Furthermore, the effect of each non-dimensional parameter on the performance of each data-driven model is investigated. Finally, the effect of number of input parameters on 28 days compressive strength of concrete is examined.
Among the unresolved issues in the design of structural concrete reinforced with
fiber reinforced composite (FRP) bars, the understanding of size effect in the reduction of the
shear strength of deep beams without shear reinforcement is of fundamental and practical significance.
Size effect accrues primarily from the larger width of diagonal cracks as the effective
depth is increased, and has been extensively documented in the case of steel reinforced concrete
(RC) through a number of laboratory tests. In FRP RC, the lower longitudinal elastic modulus
of the flexural reinforcement results in deeper and wider cracks. Yet, the calibration of any of
the current semi-empirical design algorithms is based on test results of beams and one-way slabs
with maximum effective depth of 360 mm, which is not representative of relevant large-scale
applications. This paper presents and discusses the results of laboratory testing of large-size and
scaled FRP RC beams without shear reinforcement, having maximum effective depth of 147,
294 and 883 mm, and effective reinforcement ratio of 0.12% and 0.24%. It is shown that the
shear strength of the large-size specimens with less flexural reinforcement decreases on average
by 55% compared with the smaller specimens. However, the conservativeness of the current design
algorithms generally offsets the size effect. The provisions of the UK Institution of Structural
Engineers (ISE) and the Italian National Research Council (CNR) provide the most accurate
estimates, where the former yields more conservative and consistent results.
In recent years, the use of fiber reinforced polymer (FRP) material has gained acceptance as structural reinforcement for concrete in lieu of the more extensively used steel rebar. Due to the difference in mechanical properties, FRP reinforced beams exhibit a reduction in the contribution of concrete to the overall shear resistance as compared to steel reinforced beams. This paper describes an experimental investigation of the influence of longitudinal reinforcement properties on the shear strength of beams with no transverse reinforcement. A total of 18 beams with steel, glass FRP and carbon FRP reinforcement were loaded to failure in shear. The results confirm the reduction in shear capacity with modulus of the reinforcement but also demonstrate that current suggested design equations are more conservative than experimental equations for steel reinforced beams.
To investigate the shear strength and behavior of concrete beams reinforced with fiber-reinforced polymer (FRP) bars, nine large-scale reinforced concrete beams without transverse reinforcement were tested. Three types of FRP reinforcement (two types of glass FRP and one type of aramid FRP) and two types of steel reinforcement with varying yield strengths were used in the investigation. The nominal concrete strength was 5000 psi (34.5 MPa), and the longitudinal reinforcement ratio was varied from approximately 0.36 to 2%. The specimens were simply supported and loaded with one concentrated load at midspan. The specimens were analyzed using both the ACI Committee 440 recommended shear design procedures and the ACI 318-99 shear design provisions. These results were compared with the test results. For FRP-bar-reinforced beams, the ACI 440 design method resulted in very conservative shear strength estimates, whereas the ACI 318-99 method resulted in unconservative computations of shear strength.
主筋およびせん断補強筋に連続繊維補強材を用いたコンクリートはりの実験および解析を行うことで, 連続繊維で補強したコンクリートはりのせん断耐荷力の評価を試みた. 主筋に連続繊維補強材を用いた場合には, 斜め引張破壊強度に及ぼす主筋弾性係数の影響を解析的に明らかにし, その強度評価方法を示した. 一方, せん断補強筋に連続繊維補強材を用いた場合には, 曲げ成形部の強度に注目し, その強度算定式を理論的に導くとともに, 最終的にせん断補強筋の破断で破壊に至るはりのせん断耐荷力評価方法を検討した.
Due to high strength and ductility, concrete filled steel tube columns have been highly regarded in recent decades and many experimental studies have been carried out in predicting the strength of these columns. Increase in compressive strength of concrete core by the lateral confinement provided by steel tube and delay of the steel local buckling by the contact with the hardened concrete are effective parameters in behavior of concrete filled steel tubes. This study presents a new approach to predict the capacity of circular concrete filled steel tube columns under axial loading condition, using a large number of experimental data by applying artificial neural networks. The effects of yield stress and wall thickness of steel tube, compressive strength of concrete and dimensions of column are examined. Proposed equation is compared with other existing models and indicates that the new model can predict the ultimate strength of axially loaded columns by a high level of precision.
Decision making on buildings after the earthquake have always been a great concern of scientists. Safety concerns, possibility of using the building, repairing the building, and the rate of damage are some of the most vital factors that needs to be paid attention in immediate decision makings of the buildings. In order to determine the level of damage in the buildings, the maximum displacement of stories is one of the most important parameter that needs to be investigated. In this paper, a concrete frame with shear wall containing 4-stories and 4-bays has been designed for acceleration records of 0.1 g to 1.5g and the rate of damage is determined. The total of 450 data with 6 input variables and one output variable is produced. The input parameters are defined as frequency, Vs, Richter, the distance from the earthquake epicentre (DEE), PGA, and acceleration, and the output parameter is defined as drift. With respect to this data set, three different data-driven models, i.e. artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Multiple Linear Regression Model (MLR) are used to predict the displacements. Results indicate that Both the ANN and ANFIS model show great accuracies in estimating the displacements in concrete frame with shear wall. On the other hand, MLR model did not show acceptable accuracy in the same estimation purposes. Finally, the sensitivity analysis was performed on the data set and it was observed that the accuracy of the predictions highly depends on the number of input parameters. In other words, increasing the number of input parameters would result in the increase in the accuracy of the final prediction results.
Recent tests of beams reinforced with fiber-reinforced polymer (FRP) bars indicate that traditional design methods for shear are not suitable for the design of beams incorporating these materials. The goal of this research was to develop a simple method for the design of FRP-reinforced beams and to determine if a common approach for steel and FRP-reinforced members is possible. This study presents a model for calculating the concrete contribution to shear strength of reinforced concrete beams. The applicability of the model is supported by comparing the computed shear strengths with the experimental strengths of 370 specimens. The shear strength equation developed from the model is simplified to provide a design equation that is applicable for both steel and FRP-reinforced beams. The design equation is shown to provide conservative results across the range of variables known to affect shear strength.
This paper reports experimental data on the behavior and shear strength of high-strength concrete slender beams reinforced with fiber-reinforced polymer (FRP) bars. Shear tests were conducted on six large-scale reinforced concrete beams without stirrups using high-strength concrete (f′c = 65 MPa) along with three beams using normal-strength concrete (f′c = 35 MPa). The beams measured 3250 mm long, 250 mm wide, and 400 mm deep, and were tested in four-point bending. The test variables were strength of concrete, and the reinforcement ratio and modulus of elasticity of the longitudinal reinforcing bars. Carbon and glass FRP bars and conventional steel bars were used as longitudinal reinforcement in this investigation. The experimental shear strengths of the FRP-reinforced concrete beams were compared to theoretical predictions provided by ACI 440.1R-03 and the modified form of this method proposed by the authors. The test results indicated that the high-strength concrete beams exhibited slightly lower relative shear strength compared to normal-strength concrete beams. In addition, the ACI 440.1R-03 design method provided very conservative predictions whereas the proposed modified equation gave better results.
The behavior and shear strength of concrete slender beams reinforced with fiber-reinforced polymer (FRP) bars were investigated. A total of nine large-scale reinforced concrete beams without stirrups were constructed and tested up to failure. The beams measured 3250 mm long, 250 mm wide, and 400 mm deep and were tested in four-point bending. The test variables were the reinforcement ratio and the modulus of elasticity of the longitudinal reinforcing bars. The test beams included three beams reinforced with glass FRP bars, three beams reinforced with carbon FRP bars, and three control beams reinforced with conventional steel bars. The test results were compared with predictions provided by the different available codes, manuals, and design guidelines. The test results indicated that the relatively low modulus of elasticity of FRP bars resulted in reduced shear strength compared to the shear strength of the control beams reinforced with steel. In addition, the current ACI 440.1R design method provided very conservative predictions, particularly for beams reinforced with glass FRP bars. Based on the obtained experimental results, a proposed modification to the current ACI 440.1R design equation is presented and verified against test results of other researchers.
This paper examines the effect of depth on the shear strength and behavior of high-strength concrete beams reinforced with glass-fiber-reinforced polymer (GFRP) and carbon-fiber-reinforced polymer (CFRP) bars in the longitudinal direction only without stirrups. Three beams, for each reinforcement type, with depths approximately equal to 300, 450, and 600 mm were tested to determine their shear strength and behavior before and after cracking. The targeted concrete strength was 70 MPa. The tests were carried out using two-point monotonic loading. The test results are presented in terms of crack patterns, load-deflection behavior, and failure modes. It was observed that the shear strength decreased with the increase in the depth of the beams. These results were compared with Bazant size-effect law and a good agreement was observed. The test results were also compared with the predictions using the Canadian Standard Association (CSA) and American Concrete Institute (ACI) shear design equations. The predicted results using the CSA equation were in better agreement with the experimental results than those obtained using the ACI equation. DOI: 10.1061/(ASCE)CC.1943-5614.0000248. (C) 2012 American Society of Civil Engineers.
This paper presents the results of an experimental investigation that was carried out to examine the size effect on the shear strength and behavior of concrete beams reinforced with fiber-reinforced polymer (FRP) bars. The beams were reinforced with glass FRP, carbon FRP, and steel bars in the longitudinal direction with no transverse shear reinforcement. Twelve large-scale simply supported beams with different depths, four for each reinforcement type, were tested to determine their shear capacity. The effective depth of the beams was in the range of 300-750mm. The axial stiffness of the reinforcement was kept the same for beams with the same reinforcement type with different depths. The test results indicated that as the member depth increased, the shear strength at failure decreased for all reinforcement types. The results were compared with Baant's size effect law including different parameters, and a reasonably good trend was observed. The shear strength of FRP reinforced beams were also compared with the predictions using design codes and Canadian and U.S. guideline equations. The comparisons with the equations indicated that the predicted results using one of the Canadian equations were the closest to the experimental results, while one of the U.S. equations predicted results were more conservative and gave prediction results with large scatter, especially for beams with smaller depth.
A new equation to predict the contribution of concrete to the shear strength of fiber-reinforced polymer (FRP) reinforced concrete beams without shear reinforcement is proposed. The new equation considers the elastic modulus ratio of the FRP bars to the steel reinforcement, the shear span to depth ratio, and the flexural reinforcement ratio, and was developed using the results of 60 concrete beam tests. The proposed equation more accurately predicted the results of various experiments available in the literature than the equations of an American Concrete Institute standard, and yielded similar degrees to the equations of a Canadian Standards Association standard, despite making somewhat higher predictions. The applicability of the proposed equation to FRP reinforced lightweight concrete beams was investigated using 24 all-lightweight concrete beam tests. The concrete shear strengths of the FRP reinforced all-lightweight concrete beams were equivalent to 75% of the strengths predicted by the proposed equation for normal concrete. Furthermore, with a reduction factor of 0.85, the proposed equation also showed good results for the concrete shear strength of glass FRP (GFRP) reinforced sand-lightweight concrete panels presented in a recently published paper.
An improved method for evaluating the shear resistance of fiber-reinforced polymer (FRP) reinforced concrete members without stirrups is presented. The effects of shear and moment interaction at section and of member size on its shear strength are considered. For beams with a span-depth ratio (a/d) less than 2.5, the effect of shear transfer by arch action is taken into account. Following the traditional ACI approach, the concrete contribution is defined as a function of the square root of concrete strength, but the contribution from the aggregate interlock mechanism is expressed as a function of the cubic root of the axial rigidity of longitudinal reinforcement. The member size effect on its shear strength is also considered. The predictions of the proposed method are in better agreement with available experimental data than those of any of the current shear design methods for FRP-reinforced concrete structures.
This paper evaluates the shear strength for normal and high strength concrete beams reinforced with longitudinal GFRP reinforcing bars and no web reinforcement. The effect of reinforcement ratio ρ f is examined, and experimental data is compared with values predicted by FRP shear strength expressions found in the literature, including the new design expression recommended by ACT Committee 440 (ACI 440. 2001).
This study investigates the shear behavior of concrete beams reinforced with fiber-reinforced polymer (FRP) reinforcement. Six beams were subjected to two successive phases of testing. Half of the beams were reinforced in flexure with conventional steel reinforcement, while the other half were reinforced with glass fiber bars. Different shear span to depth ratios, ranging from 1.1 to 3.3, were analyzed in order to study the variation in the shear behavior of beams characterized by different types of shear failure. No shear reinforcement was provided in the first phase of testing, while in the second phase, just enough glass and carbon shear reinforcement was provided to enable failure due to shear. The results of these tests are presented and compared to predictions according to the design recommendations proposed by the ACI and the Institution of Structural Engineers, U.K. The results of this study show that these approaches, which are based on modifications of equations derived for steel reinforcement, underestimate the contribution of the concrete and the shear reinforcement to the total shear capacity of FRP RC beams. It is shown that both approaches can be modified to become less conservative.
Seven beams were tested in bending to determine the concrete contribution to their shear resistance. The beams had similar dimensions and concrete strength and were reinforced with carbon fiber reinforced polymer bars for flexure without transverse reinforcement. They were designed to fail in shear rather than flexure. The test variables were the shear span to depth ratio, varying from 1.82 to 4.5, and the flexural reinforcement ratio, varying from 1.1 to 3.88 times the balanced strain ratio. The test results are analyzed and compared with the corresponding predicted values using the American Concrete Institute, the Canadian Standard, and the Japan Society of Civil Engineers (JSCF) fiber reinforced polymer design recommendations. Based on these results and previous experimental data, it is shown that the ACI recommendations are extremely conservative whereas the Canadian and JSCE recommendations, albeit still conservative, are in closer agreement with the experimental data. Overall the Canadian Standard's predictions are in better agreement with experimental data than the JSCE predictions.
This paper evaluates the shear strength, V-c, of intermediate length (2.5 < ald < 6) simply supported concrete beams subjected to four-point monotonic loading and reinforced with deformed, glass fiber-reinforced polymer (GFRP) reinforcement bars. Six different overreinforced GFRP designs, p > rho (b), were tested with three replicate beams per design. All samples failed as a result of diagonal-tension shear. Measured shear strengths at failure are compared with theoretical predictions calculated according to traditional steel-reinforced concrete procedures and recently published expressions intended for beams reinforced with GFRP. Recommendations are made regarding the adequacy of shear strength prediction equations for GFRP-reinforced members. The study concludes that shear capacity is significantly overestimated by the "Building Code Requirements for Structural Concrete and Commentary" (ACI 318-99) expression for, V-c, as a result of the large crack widths, small compression block, and reduced dowel action in GFRP-reinforced flexural members. Shear strength was found to be independent of the amount of longitudinal GFRP reinforcement. A simplified empirical equation for predicting the ultimate shear strength of concrete beams reinforced with GFRP is endorsed.
Corrosion of the steel reinforcement in a cold and saline environment leads to the overall deterioration of reinforced concrete structures. To avoid such deterioration, fiber-reinforced polymers (FRP) rebars are used in place of steel reinforcement. In the present paper, test results of 12 concrete beams reinforced with FRP rods are reported. The main parameters investigated in the study include the reinforcement ratio and the concrete strength. Theoretical models are proposed for the prediction of crack width, crack spacing, load-deflection response, ultimate capacity, and modes of failure. The concept of deformability is also discussed.
This paper reports test results of 12 concrete beams reinforced with glass fibre-reinforced polymer (GFRP) bars subjected to a four point loading system. All test specimens had no transverse shear nor compression reinforcement and were classified into two groups according to the concrete compressive strength. The main parameters investigated in each group were the beam depth and amount of GFRP reinforcement. Two modes of failure were observed, namely flexural and shear. The flexural failure is mainly occurred due to tensile rupture of GFRP bars either within the mid-span region or under the applied point load. The shear failure is initiated by a major diagonal crack within the beam shear span. This diagonal crack extended horizontally at the level of the GFRP bars indicating bond failure.Simplified methods for estimating the flexural and shear capacities of beams tested are presented. The flexural capacity is estimated based on the compatibility of strains and equilibrium of forces. Comparisons between the flexural capacity obtained from the theoretical analysis and that experimentally measured in the current investigation and elsewhere show good agreement. To predict the shear capacity of the beams tested, four methods recently proposed in the literature for GFRP-reinforced concrete beams are used. These methods have been developed by modifying the ACI 318-99 shear capacity formula for steel-reinforced concrete beams to account for the difference in the axial stiffness of GFRP and steel bars. It has been shown that the theoretical predictions of the shear capacity obtained from these methods are inconsistent and further research needs to be carried out in order to establish a rational method for the shear capacity calculation of GFRP-reinforced concrete beams.
Fiber Reinforced Plastic (FRP) reinforcements are currently used for special concrete structures in areas sensitive to magnetic fields and severe environmental conditions which accelerate corrosion of the steel reinforcements, and consequently leads to deterioration of the structure. This paper presents test results of eight one-way concrete slabs reinforced with glass-fiber, carbon-fiber and conventional steel reinforcements. The slabs were tested under static loading conditions to determine their flexural and shear limit states, including the behavior prior to cracking, cracking, ultimate capacities and modes of failure. Based on this investigation, design recommendations and guidelines are proposed.
The surface-water hydrographs of rivers exhibit large variations due to many natural phenomena. One of the most commonly used approaches for interpolating and extending streamflow records is to fit observed data with an analytic power model. However, such analytic models may not adequately represent the flow process, because they are based on many simplifying assumptions about the natural phenomena that influence the river flow. This paper demonstrates how a neural network can be used as an adaptive model synthesizer as well as a predictor. Issues such as selecting an appropriate neural network architecture and a correct training algorithm as well as presenting data to neural networks are addressed using a constructive algorithm called the cascade-correlation algorithm. The neural-network approach is applied to the flow prediction of the Huron River at the Dexter sampling station, near Ann Arbor, Mich. Empirical comparisons are performed between the predictive capability of the neural network models and the most commonly used analytic nonlinear power model in terms of accuracy and convenience of use. Our preliminary results are quite encouraging. An analysis performed on the structure of the networks developed by the cascade-correlation algorithm shows that the neural networks are capable of adapting their complexity to match changes in the flow history and that the models developed by the neural-network approach are more complex than the power model.
Increasing interest in the use of fiber-reinforced polymer (FRP) reinforcement for reinforced concrete structures has made it clear that insufficient information about the shear performance of such members is currently available to practicing engineers. This paper summarizes the results of 11 large shear tests of reinforced concrete beams with glass FRP (GFRP) longitudinal reinforcement and with or without GFRP stirrups. Test variables were the member depth, the member flexural reinforcement ratio, and the amount of shear reinforcement provided. Results showed that the equations of the Canadian CSA shear provisions provide conservative estimates of the shear strength of FRP-reinforced members. Recommendations are given along with a worked example on how to apply these provisions including to members with FRP stirrups. It was found that members with multiple layers of longitudinal bars appear to perform better than those with a single layer of longitudinal reinforcing bars. Overall, it was concluded that the fundamental shear behavior of FRP-reinforced beams is similar to that of steel-reinforced beams despite the brittle nature of the reinforcement.
Strengthening and retrofitting of concrete columns by wrapping and bonding FRP sheets has become an efficient technique in recent years. Considerable investigations have been carried out in the field of FRP-confined concrete and there are many proposed models that predict the compressive strength which are developed empirically by either doing regression analysis using existing test data or by a development based on the theory of plasticity. In the present study, a new approach is developed to obtain the FRP-confined compressive strength of concrete using a large number of experimental data by applying artificial neural networks. Having parameters used as input nodes in ANN modeling such as characteristics of concrete and FRP, the output node was FRP-confined compressive strength of concrete. The idealized neural network was employed to generate empirical charts and equations for use in design. The comparison of the new approach with existing empirical and experimental data shows good precision and accuracy of the developed ANN-based model in predicting the FRP-confined compressive strength of concrete.
Fibre reinforced polymer (FRP) bars represent an interesting alternative to conventional steel as internal reinforcement of reinforced concrete (RC) members where some properties such as durability, magnetic transparency, insulation, are of primary concern. The present paper focuses on the assessment of Eurocode-like design equations for the evaluation of the shear strength of FRP RC members, as proposed by the guidelines of the Italian Research Council CNR-DT 203 [CNR-DT 203/2006. Guide for the design and construction of concrete structures reinforced with fiber-reinforced polymer bars. National Research Council, Rome, Italy; 2006]. Both the concrete and the FRP stirrups contributions to shear are taken into account: the new equations derived with reference to Eurocode equations for shear of steel RC members are verified through comparison with the equations given by ACI, CSA and JSCE guidelines, considering a large database of members with and without shear reinforcement failed in shear.
A proper design of the architecture of Artificial Neural Network (ANN) models can provide a robust tool in water resources modeling and forecasting. The performance of different neural networks in a groundwater level forecasting is examined in order to identify an optimal ANN architecture that can simulate the decreasing trend of the groundwater level and provide acceptable predictions up to 18 months ahead. Messara Valley in Crete (Greece) was chosen as the study area as its groundwater resources have being overexploited during the last fifteen years and the groundwater level has been decreasing steadily. Seven different types of network architectures and training algorithms are investigated and compared in terms of model prediction efficiency and accuracy. The different experiment results show that accurate predictions can be achieved with a standard feedforward neural network trained with the Levenberg–Marquardt algorithm providing the best results for up to 18 months forecasts.
The use of fibre reinforced polymer (FRP) bars to reinforce concrete structures has received a great deal of attention in recent years due to their excellent corrosion resistance, high tensile strength, and good non-magnetization properties. Due to the relatively low modulus of elasticity of FRP bars, concrete members reinforced longitudinally with FRP bars experience reduced shear strength compared to the shear strength of those reinforced with the same amounts of steel reinforcement. This paper presents a simple yet improved model to calculate the concrete shear strength of FRP-reinforced concrete slender beams (a/d>2.5) without stirrups based on the gene expression programming (GEP) approach. The model produced by GEP is constructed directly from a set of experimental results available in the literature. The results of training, testing and validation sets of the model are compared with experimental results. All of the results show that GEP is a strong technique for the prediction of the shear capacity of FRP-reinforced concrete beams without stirrups. The performance of the GEP model is also compared to that of four commonly used shear design provisions for FRP-reinforced concrete beams. The proposed model produced by GEP provides the most accurate results in calculating the concrete shear strength of FRP-reinforced concrete beams among existing shear equations provided by current provisions. A parametric study is also carried out to evaluate the ability of the proposed GEP model and current shear design guidelines to quantitatively account for the effects of basic shear design parameters on the shear strength of FRP-reinforced concrete beams.
Guide for the Design and Construction of Structural Concrete Reinforced with FRP Bars
- American Concrete Institute
American Concrete Institute (ACI) Committee 440, Guide for the Design and
Construction of Structural Concrete Reinforced with FRP Bars, ACI 440.1R-06,
American Concrete Institute, Farmington Hills (MI), 2006, p. 44.
Reinforcing Concrete Structures with Fiber Reinforced Polymers, ISISM03-07, Canadian network of Centers of Excellence on Intelligent Sensing for Innovative Structures
- I S I S Canada
I.S.I.S. Canada, Reinforcing Concrete Structures with Fiber Reinforced Polymers,
ISISM03-07, Canadian network of Centers of Excellence on Intelligent Sensing for
Innovative Structures, University of Winnipeg, Manitoba, 2007, p. 151.
Interim Guidance on the Design of Reinforced Concrete Structures using Fiber Composite Reinforcement
- Istructe
- Seto Ltd
British Institution of Structural Engineers (BISE), Interim Guidance on the Design of
Reinforced Concrete Structures using Fiber Composite Reinforcement, IStructE,
SETO Ltd., London, 1999.
Recommendation for design and construction of concrete structures using continuous fiber reinforcing materials
Japan Society of Civil Engineers, JSCE., Recommendation for design and construction of concrete structures using continuous fiber reinforcing materials, in: A.
Machida (Ed.), Concrete Engineering Series, vol. 23, Tokyo, Japan, 1997, p. 325.
Shear capacity of RC and PC beams using FRP reinforcement
- Tottori
S. Tottori, H. Wakui, Shear capacity of RC and PC beams using FRP reinforcement,
ACI Spec. Publ. 138 (1993) 615-632.
Shear performance of concrete beams reinforced with FRP stirrups
- T Nagasaka
- H Fukuyama
- M Tanigaki
T. Nagasaka, H. Fukuyama, M. Tanigaki, Shear performance of concrete beams
reinforced with FRP stirrups, ACI Spec. Publ. 138 (1993) 789-812.
Structural implications of using GFRP bars as concrete reinforcement
- N Swamy
- M Aburawi
N. Swamy, M. Aburawi, Structural implications of using GFRP bars as concrete
reinforcement, in: Proc. Third Int. Symp. Non-metallic Reinf. Concr. Struct., vol. 3,
1997, pp. 503-510.
Influence of Different Parameters on Shear Strength of FRP Reinforced Concrete Beams without Web Reinforcement
- M Alam
M. Alam, Influence of Different Parameters on Shear Strength of FRP Reinforced
Concrete Beams without Web Reinforcement, University of Newfoundland, 2010.
Shear performance of concrete beams reinforced with FRP stirrups
- Nagasaka