Steel Fiber Reinforced Concrete Research Articles

Efficacious application of data-driven machine learning models for predicting and optimizing the flexural tensile strength of fiber-reinforced concrete

In the literature, a limited number of studies have utilized the random forest (RF) technique to model the flexural tensile strength (FTS) of steel fiber reinforced concrete (SFRC). This study established modeling of the FTS of SFRC containing mono-fibrous (exclusively crimped and hooked) systems. A comprehensive database of 193 records was used from 18 published studies (including 30 documents sampled by authors). Machine-learning numerical models were developed and evaluated. Partial dependence plots were constructed to visualize the relationship between the model parameters and rank the parameters based on their importance. It has been found that the RF model can be rationally used to predict and optimize the FTS of SFRC, as most of the predicted–tested results were close to the ± 80% accuracy range. The model showed of underestimation nature for SFRC with high FTS values. The Gini index-based importance analysis showed that the fiber’s content and tensile strength had the highest significance among the other variables. A drastic increase in the FTS by an increase in the fibrous dosage was achieved. However, no major SFRC’s tensile strength enhancement for dosages was more than 1.5%, while its optimum diameter was 0.5–0.7 mm. The results also illustrated that the effect of the fiber’s length is quite similar to that of the fiber content. However, no significant increase can be achieved by increasing the fiber’s length from 40 to 60 mm.

Stress-Strain State of Steel Fiber-Reinforced Concrete under Compression Taking into Account Unloading from Inelastic Region

The purpose of the study is to examine the physical and mechanical characteristics of steel fiber-reinforced concrete under compression, including: modulus of elasticity, Poisson ratio, values of ultimate strains under compression, values of compressive strength with different percentages of dispersed reinforcement. An experimental investigation program, which included the production of cube samples measuring 100×100×100 mm, as well as a compression test under static loading, taking into account unloading from the region of inelastic deformations, was developed and carried out. Two types of steel fiber were chosen as dispersed reinforcement: hooked end and wave shape. The volume content of steel fiber in the cube samples was 0.5, 1.0, 1.5 and 2.0 %. As a result of the investigation, the strength and deformation characteristics of steel fiber reinforced concrete under compression were obtained. Based on the experimental data, actual strain diagrams of steel fiber reinforced concrete were constructed, taking into account the type of reinforcing fibers and the percentage of reinforcing fiber. Based on the obtained diagrams, a law of deformation of steel fiber reinforced concrete is proposed, which can be described by a polynomial function of the fourth order with constant coefficients that determine the shape of the stress-strain curve. The presented research results can be used in developing a methodology for physically nonlinear analysis of steel fiber reinforced concrete elements with a percentage of dispersed reinforcement from 0.5 to 2.0 %.

Assessing sampling interval-dependent roughness in fractured steel fiber reinforced concrete using a double-exponential decay model

This study introduces a double-exponential decay (DED) model to investigate the sampling interval (δ)-dependent behavior of crack morphology roughness in steel fiber reinforced concrete (SFRC). The root mean square (Z2-3D) of the directional derivatives in 3D digital morphology is employed to quantify the roughness and its anisotropic behavior. Surface morphologies of 24 fractured specimens are captured by using a laser scanner and the obtained results are re-meshed based on the Point-Kriging algorithm. The roughness is represented by the ratio of surface area to projected area (RS) and Z2-3D values along different directions at various δs (4–3000 μm). The outcomes suggest that the values of RS and Z2-3D both decline, but the anisotropy of surface roughness is enhanced at an increased δ. The DED model is able to properly reflect the decay behavior in which the morphology roughness is classified into three corrugated components based on their distinct attenuation laws. A satisfactory agreement with the literature data is observed, validating the presented methodology. The reported findings in this work are useful for enhancing the understanding of the formation of crack morphology roughness and its relationship to the damage process of SFRC.

Predicting the fire-induced structural performance of steel tube columns filled with SFRC-enhanced concrete: using artificial neural networks approach

Predicting the axial Shortening strength of concrete-filled steel tubular (CFST) columns is an important problem that this study attempts to solve for civil engineering projects. We suggest using a deep learning-based artificial neural network (ANN) model to address this issue, taking into account the intricate relationship between steel tube and core concrete. The model, called ANN-SFRC (Steel Fibre Reinforced Concrete), surpasses an R2 threshold of 0.90 and achieves impressive R2 values across different types of CFST columns. Compared to traditional linear regression methods, the ANN-SFRC model significantly improves accuracy, with an observed inaccuracy of less than 3% compared to actual values. With its reliable approach to forecasting the behavior of CFST columns under axial compression, this high-performance instrument enhances safety and accuracy during the design and planning stages of civil engineering.

Low-velocity impact behavior of two-way SFRC slabs strengthened with steel plate

Structural systems and structural elements can often suffer severe damage or even completely collapse under the effect of sudden dynamic impact loading, which is a different type of loading that is not considered during their design. Research on how structures behave under impact loading and how they can be strengthened to perform better against this type of loading has increased to avoid such undesirable severe damage. Within the scope of this study, it is aimed to improve the behavior and increase the performance of two-way steel fiber-reinforced concrete (SFRC) slabs, one of the leading structural elements that can be affected by impact loading, using steel fiber concrete (SFC) and placing steel plates on the surface of the RC slab. Within the scope of the study, the effects of placing FRC as layers in different positions within the slab and placing the steel plate on different surfaces of the slabs were examined. Impact loading was applied using a drop weight test setup designed by the authors, and the acceleration–time, displacement–time, and impact loading–time behaviors of the RC slabs were measured and interpreted. The use of fiber concrete in RC slabs and strengthened with steel plates increased the maximum acceleration values by an average of 3% and 113%, respectively. The use of fiber concrete in RC slabs reduced the maximum displacement and residual displacement values by an average of 2% and 25%, respectively. Placing steel plates on the slabs reduced the maximum displacement and residual displacement values by an average of 270% and 199%, respectively. In addition, the energy absorption capacities of RC slabs were calculated, and how they were affected by experimental variables was examined. Numerical analyses of the RC slabs tested in the study were also conducted using ABAQUS finite element software, and the results obtained were compared with the experimental ones.

RECYCLING METHODS FOR STEEL FIBER REINFORCED CONCRETE

The favorable impact of steel fibers in concrete is well known. Steel fibers can affect various fresh and hardened properties of concrete, for instance the appearance and propagation of cracks in the concrete matrix. Thus far, extensive research has been conducted to describe and to analyze those phenomena. Furthermore, Steel Fiber Reinforced Concrete (SFRC) is state of the art and is used extensively in industrial floors, walls, and ceiling constructions. But, beside the beneficial impact, the recycling of SFRC and, in particular, the separation of crushed concrete and steel fibers should be focused too. Only a few research programs are described in literature and comprehensive suggestions or guidelines are missing completely. So, a first test program was carried out to evaluate the usual procedure to break the concrete with a jaw crusher and to separate the contained steel via magnet. In tests, the amount and the type of the fibers used were varied as well as the concrete strength. The tests suggest that a certain amount of fibers can be separated, but the majority of fibers is still bound to the crushed concrete. Via sieving, it is possible to further separate concrete still containing fibers from aggregates without fibers. Conglomerates of broken concrete and fibers may be crushed again afterwards, but, allegedly, the success of the described procedure strongly depends on the fiber amount and the concrete properties. Furthermore, the available test data is too small to sufficiently describe the entire recycling path, making further research inevitable.

RECYCLING METHODS FOR STEEL FIBER REINFORCED CONCRETE

The favorable impact of steel fibers in concrete is well known. Steel fibers can affect various fresh and hardened properties of concrete, for instance the appearance and propagation of cracks in the concrete matrix. Thus far, extensive research has been conducted to describe and to analyze those phenomena. Furthermore, Steel Fiber Reinforced Concrete (SFRC) is state of the art and is used extensively in industrial floors, walls, and ceiling constructions. But, beside the beneficial impact, the recycling of SFRC and, in particular, the separation of crushed concrete and steel fibers should be focused too. Only a few research programs are described in literature and comprehensive suggestions or guidelines are missing completely. So, a first test program was carried out to evaluate the usual procedure to break the concrete with a jaw crusher and to separate the contained steel via magnet. In tests, the amount and the type of the fibers used were varied as well as the concrete strength. The tests suggest that a certain amount of fibers can be separated, but the majority of fibers is still bound to the crushed concrete. Via sieving, it is possible to further separate concrete still containing fibers from aggregates without fibers. Conglomerates of broken concrete and fibers may be crushed again afterwards, but, allegedly, the success of the described procedure strongly depends on the fiber amount and the concrete properties. Furthermore, the available test data is too small to sufficiently describe the entire recycling path, making further research inevitable.

A SIMPLE APPROACH FOR PREDICTING CURVATURE IN CONCRETE MEMBERS REINFORCED WITH STEEL FIBERS AND BARS

The paper proposes a simple deflection model for concrete bending members reinforced with steel fibers and bars (further denoted as R/SFRC members). Most of the current design approaches for such elements require the knowledge of the residual strength of steel fiber reinforced concrete (SFRC) in tension with the latter characteristic being obtained from standard bending tests. The involvement of extra tests hampers a wider application of R/SFRC structures in practice. The proposed model does not require the characteristic of the residual strength. It was derived using an inverse approach earlier proposed by the authors. The resultant tensile force of SFRC acting in the tension zone of a R/SFRC beam was derived from the experimental moment-curvature diagram. A load level was identified at which the resultant tensile force reached zero, implying that the section stiffness equals to the stiffness of the fully cracked RC section. The proposed model in a simple way expresses curvature via the curvatures at a cracking load and the load corresponding to the stiffness of the fully cracked RC section. The predictions by the proposed model were checked against independent test data.

A SIMPLE APPROACH FOR PREDICTING CURVATURE IN CONCRETE MEMBERS REINFORCED WITH STEEL FIBERS AND BARS

The paper proposes a simple deflection model for concrete bending members reinforced with steel fibers and bars (further denoted as R/SFRC members). Most of the current design approaches for such elements require the knowledge of the residual strength of steel fiber reinforced concrete (SFRC) in tension with the latter characteristic being obtained from standard bending tests. The involvement of extra tests hampers a wider application of R/SFRC structures in practice. The proposed model does not require the characteristic of the residual strength. It was derived using an inverse approach earlier proposed by the authors. The resultant tensile force of SFRC acting in the tension zone of a R/SFRC beam was derived from the experimental moment-curvature diagram. A load level was identified at which the resultant tensile force reached zero, implying that the section stiffness equals to the stiffness of the fully cracked RC section. The proposed model in a simple way expresses curvature via the curvatures at a cracking load and the load corresponding to the stiffness of the fully cracked RC section. The predictions by the proposed model were checked against independent test data.

Study on the evolution patterns of concrete lining’s physical and mechanical properties under sulfuric acid corrosion conditions

During operation, deep water storage and drainage tunnels may face sulfuric acid corrosion resulting from microbiologically induced chemical reactions. In this study, sulfuric acid corrosion experiments were performed on C60 concrete specimens with steel fiber volume ratios of 0% (plain) and 1% using a sulfuric acid solution with an initial pH of 1. For one-face corrosion, five faces of cubic concrete samples were coated with exposed resin, leaving only one face exposed to sulfuric acid. For all-face corrosion, all faces were exposed to sulfuric acid. Concrete corrosion under different loading conditions was compared by adding weights on top of the concrete. Rate of mass change variations and uniaxial compressive strength at various corrosion time points were monitored. The results indicated that after sulfuric acid corrosion, there was an increase in the mass but an overall decrease in the strength of the specimens. Compared with all-face corrosion, one-face corrosion led to a smaller increase in the mass of the concrete specimens and a more substantial decrease in strength, primarily because of the slower corrosion rate and reduced gypsum formation. Compared to plain concrete, steel fiber-reinforced concrete experienced a lower increase in the mass of the concrete specimens because the steel fibers obstructed sulfuric acid infiltration. However, there was significant corrosion of the steel fibers and surrounding concrete owing to accumulation of sulfuric acid around the former, resulting in a more substantial decrease in the uniaxial compressive strength of the steel fiber-reinforced concrete. With increasing load, the rate of change in the concrete mass initially decreased and then increased; this could be because increased loads densify the concrete, hindering sulfuric acid infiltration and promoting the development of microcracks within the concrete, thereby enhancing the effects of sulfuric acid corrosion.

Design and flexural behavior of steel fiber-reinforced concrete beams with regular oriented fibers and GFRP bars

In this study, steel fiber-reinforced concrete (SFRC) beam structures with regular oriented steel fibers are designed and prepared by using an assembled solenoid device. Fiber orientation is not arranged in a single direction but is similar to the direction of tensile stress of the beam in bending load conditions. With application of inductive test method and image processing technology, fiber orientation coefficient of SFRC was evaluated and increased from 0.5 to 0.7–0.8 after fiber alignment. To describe and analyze the flexural behavior of SFRC beams with regular oriented fibers and glass fiber-reinforced polymer (GFRP) rebars, 14 concrete beams with different parameters were fabricated and subjected to a four-point bending test. Specifically, the parameters included fiber content, fiber orientation, GFRP reinforcement ratio and different layer sections. The results showed that optimizing fiber orientations to the tension zone of the section led to a higher flexural strength and can well restrain the development of crack widths with comparison to randomly distributed fibers. The peak load and energy absorption increased by 7 % and 12.4 % after fiber alignment with a fiber volume fraction of 1.5 %. The combination of randomly and regularly distributed steel fibers in the form of two-layer composite members can be an efficient solution to improve the flexural performance for purpose of the optimum distribution and direction of steel fibers, and reduce the engineering cost.

Failure characteristics and seismic behavior of steel basalt hybrid fiber reinforced concrete lining for the tunnel in strong earthquake areas

The effect of a high-intensity earthquake on a tunnel can be catastrophic. To minimize the damage of the earthquake to the tunnel, the mechanical properties of tunnel lining need to be improved in strong earthquake areas. Thereby, this paper proposes a steel basalt hybrid fiber reinforced concrete (SBHFRC) as an alternative construction material for the lining. Then, the failure characteristics and seismic performance of SBHFRC are studied and compared with that of plain concrete (PC) and steel fiber reinforced concrete (SFRC). The results show that the toughness of SFRC is better than that of SBHFRC. Furthermore, the initial crack load and failure load of SBHFRC lining are significantly higher than that of PC lining. The numbers of cracks observed in the SBHFRC lining and SFRC lining are more than that in the PC lining. But the distribution of cracks in SBHFRC lining is relatively uniform. The bearing capacity of SFRC lining and SBHFRC lining is basically the same, although significantly higher than that of PC lining. The seismic performance of different fiber concrete linings is further explored through numerical simulation. Under strong earthquakes, the damage areas of the lining are mainly concentrated at the tunnel vault bottom and vault near the fault, and the SBHFRC damage degree and range are the smallest. Compared with PC lining, the minimum safety factors of SFRC lining and SBHFRC lining are increased by 7.5 times and 11.2 times, respectively, which shows that SBHFRC lining has good seismic performance.

Effect of steel fibre with different orientations on mechanical properties of 3D-printed steel-fibre reinforced concrete: Mesoscale finite element analysis

It is widely recognized that steel fibres exhibit directional distribution during the preparation of 3D-printed steel fibre reinforced concrete (SFRC). The degree of fibre orientation varies due to several factors, such as nozzle size and layer height. This variation in fibre orientation range may result in different mechanical performance exhibited by 3D-printed SFRC at hardened state. To explore the relationship between steel fibre orientation and the mechanical properties of 3D-printed SFRC at hardened state, this study firstly establishes three-dimensional mesoscale finite element models based on the distribution of fibres in 3D-printed SFRC. The steel fibre and matrix work together through a mechanism for fluid-structure interaction between models. Then, compares the simulation results with experimental data to verify the accuracy of the models. After that, different steel fibre distribution orientations were set, and parametric analysis was conducted for the compressive and tensile loading conditions to explore the effects of fibre orientation on the damage mode and strength of 3D-printed SFRC. The results indicate that 3D-printed SFRC exhibits different damage modes with varying steel fibre orientation ranges. Additionally, the strength of 3D-printed SFRC varies in steel fibre orientation range and exhibits anisotropic characteristics. This study provides a theoretical basis for improving and controlling the performance of 3D-printed SFRC in future engineering applications.

A Novel Repair Technique of Pre-damaged T-Beams Failing in Shear Using Eco-Friendly Steel Fibre-Reinforced Geopolymer Concrete

Repair of reinforced concrete structures is required to preserve the adequate performance of these structures throughout their service life. One of the credible techniques is using fibrous concrete as a repair material. In this paper, the performance of steel fibre-reinforced geopolymer concrete (SFRGPC) in the repair of pre-damaged reinforced concrete T-beams (pre-loaded up to 50% of their shear capacity) failing in shear was investigated. Five T-beam series and a four-point loading test were adopted: one reference beam, three beams were repaired with different fibrous ratios of 1, 2, and 3%, and one was repaired with 2% steel fibre and additional U-steel stirrups. The key test results include crack propagation, crack width, initial stiffness, load deflection, peak loads, and strain associated with web stirrups. A clear enhancement was noticed in the performance of the repaired T-beams; their shear capacity was boosted by as much as 45% compared to the control beam. It was also deduced that the beam went from a brittle to a ductile failure mode at 3% SFRGPC and at 2% SFRGPC with U-stirrups. Finally, an analytical model prediction was proposed to predict the shear capacity of repaired T-beams with the SFRGPC. The model showed a satisfactory correlation with experimental results, with an average ratio of 0.995 and a standard deviation of 0.035.

Experimental investigation on axial compressive and splitting tensile behaviour of reinforced concrete with a low content of hybrid steel fibres

This study intends to explore how the hybrid steel fibre reinforcement with a low content affects the mechanical characteristics of concrete. The axial compression properties of hybrid steel fibre-reinforced concrete (HSRC) were systematically investigated regarding stress-strain curves, failure modes, elastic modulus, and toughness. The crack propagation and fracture process of HSRC were analyzed based on the cumulative crack width and horizontal strain field. The results showed that the micro-straight steel fibre (mSF) demonstrated a more pronounced influence on post-peak stress retention in the axial compression test, whereas macro hooked-end steel fibre (MHF) positively affected the peak strain enhancement. The average peak stress of the HSRC increased by 8.6 % higher than that of H40 and 16.4 % higher than that of S40. It is noteworthy that the mixing of steel fibres at a low fibre content was detrimental to the compressive toughness. The splitting tensile test showed a transitional stage before the HSRC reached its peak load, during which the load steadily increased and crack propagation slowly occurred. Moreover, the digital image correlation (DIC) measurement results suggested that mSF can enhance the internal quality of concrete. The study contributes to a deep understanding of the synergistic reinforcement mechanism and the fracture processes in HSRC.

A numerical critical shear crack model and its application to post‐peak behavior assessment of RC and SFRC beams

AbstractIn this paper, a numerical critical shear crack model, in which the shear‐flexural coupling effect is considered and the full process of stress and strain is captured, is proposed to evaluate the post‐peak behavior of reinforced concrete (RC) and steel fiber reinforced concrete (SFRC) beams. The shape of critical shear crack is identified by combining the modified compression field theory considering the bridge effect of steel fibers and the section analysis method. Based on the shape of critical shear crack, the shear capacity of beam members is provided by the shear tension zone and the shear compression zone. The shear capacity of the shear tension zone is calculated by the modified compression field theory and that of the shear compression zone is determined by multiaxial strength criterion of concrete. The vertical displacements caused by the flexure deformation and shear deformation are deduced by the moment area method and integration of shear strain. To verify the proposed numerical approach, a test database of 486 RC beams and 313 SFRC beams was established to predict the shear strength, and the force‐displacement relationships of twelve beam members are used to validate the feasibility for full process analysis. The stochastic analysis of beams with different failure modes is conducted via GF‐discrepancy‐based point selection method and probability density evolution method. The limitation between different failure modes is defined according to the degradation percent of shear capacity and it is taken as threshold value of failure domain, and the failure probability analysis indicates that the designed flexure beam member suffered severe degradation of shear capacity, resulting in a significant decline in safety probability.

Design and economic implications of using steel fibers in elevated slabs of multi‐story buildings

AbstractThis paper presents a comprehensive comparison between conventional reinforced concrete (RC) and steel fiber‐reinforced concrete (SFRC) slabs. Twenty‐two‐way rectangular RC and SFRC slabs are designed with varying dimensions, aspect ratios, and boundary conditions, following verified international standards to compare the steel quantities in a unique study. SFRC slabs with larger aspect ratios and discontinuous edges required significantly higher steel percentages than square slabs, as the random three‐dimensional distribution of steel fibers suited the load distribution pattern of square slabs. To further conduct a detailed design and economic feasibility case study on a realistic building model, two 12‐story buildings are computationally designed with RC and SFRC alternatives. The results indicate that SFRC slabs required approximately 25% more steel than RC slabs due to the uniform distribution of steel fibers throughout the section. Project cost and time estimates were conducted based on standard productivity manuals and cost databases in two distinct economies, India and the USA. Despite the higher steel requirements, the 12‐story SFRC building showed time savings of 34.1 days (5.44%) and cost savings of 4.88% and 1.27% when constructed in the USA and India, respectively. The results are compared with limited available data in literature. In conclusion, the study suggests that SFRC slabs in multi‐story buildings could be more economically favorable in countries with higher labor costs.

Application of metaheuristic algorithms for compressive strength prediction of steel fiber reinforced concrete exposed to high temperatures

The demand for steel fiber-reinforced concrete (SFRC) in construction has surged, particularly due to its enhanced fire resistance, leading to extensive research on its residual properties after elevated temperature exposure. However, conducting experimental tests can be a time-consuming process, demanding significant resources in terms of time, cost, and human resources. In contrast, machine learning (ML) models can rapidly simulate outcomes, enabling researchers to explore a wide range of scenarios efficiently. The previous literature studies used either ensemble or individual models; however, this study made a unique approach to utilize hybrid models, which are more precise compared to the other types of ML models. Accordingly, a data-rich framework containing 304 data samples was utilized in this study to develop an efficient and robust predictive model for the compressive strength (CS) of SFRC at higher temperatures. Support vector regression (SVR) in combination with three distinct optimization algorithms, specifically, particle swarm optimization (PSO), the firefly algorithm (FFA), and grey wolf optimization (GWO) were utilized to develop the hybrid models. In addition, typical ML techniques such as random forest (RF) and decision tree (DT) were used for comparative analysis. Among the five models, the SVR-FFA hybrid model showcased superior predictive accuracy, with the SVR-GWO model following closely. The correlation coefficients (R) for both models exceeded 0.99. The SVR-FFA model demonstrated relative root mean square error (RMSE) and mean absolute error (MAE) values of 0.0375 and 0.0248, respectively, while the SVR-GWO model exhibited corresponding values of 6.0560 and 5.1326. The SHapley Additive exPlanation (SHAP) technique revealed that the influence of temperature is more significant compared to the heating rate. Moreover, a considerable enhancement was observed in CS as the steel fiber volume fraction (Vf) reached 1.5%; on the other hand, no further improvement was noticed when Vf exceeded the 1.5% threshold. The proposed hybrid models are robust and accurate methods to estimate the CS of SFRC in field applications.

Structural Behavior of Circular Concrete Columns Reinforced with Longitudinal GFRP Rebars under Axial Load

This paper presents experimental and theoretical assessments of the structural behavior of circular steel fiber-reinforced concrete (SFRC) columns reinforced with glass fiber-reinforced polymer (GFRP) bars subjected to a concentric axial compressive load. Laboratory experiments were planned to evaluate and compare the effect of different design parameters on the structural behavior of column specimens based on experiments and finite element (FE) analysis. The experimental variables were (i) concrete types, i.e., conventional concrete (CC) and fiber-reinforced concrete (FC), (ii) longitudinal reinforcement types, i.e., steel and GFRP bars, and (iii) transverse rebar configurations, i.e., tied and spiral with different pitches. Sixteen column specimens were fabricated and categorized into four groups with respect to rebar configurations and concrete types. The results showed that the failure modes and cracking patterns of those four column groups were comparable, particularly in the pre-peak branches of load-deflection curves. Even though the average ultimate load of the columns with longitudinal GFRP bars was 17.9% less than that with longitudinal steel bars, the ductility index (DI) was 10.2% greater than their counterpart on average. The addition of steel fibers (SF) to concrete increased the axial peak load by up to 3.1% and the DI by up to 6.6% compared to their counterpart CC columns without SFs. The DI of specimens was increased by higher volumetric ratios (up to 12%) and spiral types (up to 5.5%). The concrete damage plastic (CDP) model for FC columns was updated in the finite element software ABAQUS 6.14. Finally, a new simple equation was theoretically proposed to predict the axial capacity of specimens by considering the inclusion of longitudinal GFRP rebars, volumetric ratio, and steel spiral/hoop ties. Good agreement between the proposed model predictions and the experimental/numerical results was observed.

A comprehensive comparison of various machine learning algorithms used for predicting the splitting tensile strength of steel fiber-reinforced concrete

Incorporating steel fibers into concrete effectively enhances its tensile strength by managing crack formation, increasing toughness and ductility, and reducing brittleness, making it a valuable choice for various applications, including bridge decks. Thus, the objective of this research is to employ the machine learning (ML) technique to estimate the splitting tensile strength (STS) of steel fiber-reinforced concrete (SFRC) that incorporates hooked industrial steel fibers (ISF) using SFRC mixtures collected from various research papers in this field. The dataset comprises 116 data points from various research papers. Utilizing the recursive feature elimination with cross-validation and leave-one-out (RFECV-LOO) method, the most influential parameters, including ISF content, water-to-cement ratio (W/C), cement content (C), fine aggregate content (FA), superplasticizer (SP), and fly ash, were identified to train the ML models. Eight statistical metrics, including the root mean squared error (RMSE), mean squared error (MSE), mean absolute error (MAE), coefficient of determination (R2), mean absolute percentage error (MAPE), mean bias error (MBE), statistical test (Tstat), and scatter index (SI) were employed as metrics to assess model performance. Various ML models including multiple linear regression (MLR), ridge regression (Ridge), lasso regression (Lasso), elastic net regression (ElasticNet), k-nearest neighbor (KNN), support vector regression (SVR), categorical boosting (CatBoost), light gradient boosting model (LightGBM), random forest (RF), gradient boosting (GB), extreme gradient boosting (XGB), adaptive boosting (AdaBoost), and multilayer perceptron (MLP) were employed to predict the SFRC’s STS. The Python programming language, TensorFlow framework as well as the Scikit-learn packages, were used to run the models. Among the various ML methods, SVR demonstrated higher accuracy, achieving an R2 of 0.930, RMSE of 0.502, and MAE of 0.330. In contrast, linear regression models, especially Lasso with an R2 of 0.631, RMSE of 0.854, and MAE of 0.710, exhibited the least effective performance when it came to predicting the STS of SFRC. Moreover, the performance of the developed ML models was approved through sensitivity analysis as well as parametric analysis.