Rectangular Concrete-filled Steel Tube Research Articles

Experimental and analytical study on the dynamic response of rectangular concrete-filled steel tube under transverse impact loading

Rectangular concrete-filled steel tubes (CFSTs) have been widely employed as prominent structural members in modern constructions. During their life cycles, rectangular CFST members may suffer from impact loads, leading to a need for investigations into their dynamic performance. In this context, 23 rectangular CFSTs subjected to transverse impact loading were experimentally investigated using a drop weight testing system. The instantaneous impact force responses and motion states of the specimens were recorded at high frequencies. The aspect ratio, the steel ratio and the clear span of the specimen, as well as the impact energy conditions, were the key parameters in the test scheme, and their influence on the dynamic response of the rectangular CFST specimens was analyzed. A finite element (FE) model was developed to predict the impact response of rectangular CFSTs, in which the strain rate effects of steel and concrete were well considered, and good accuracy was achieved. Then, an equivalent relationship between rectangular and square CFSTs under transverse impact was proposed and verified, which achieves access for rectangular CFSTs to existing design methods.

Natural Vibration Characteristics for Prestressed Concrete-Filled Rectangular Steel Tube Simply Supported Beam

Natural Vibration Characteristics for Prestressed Concrete-Filled Rectangular Steel Tube Simply Supported Beam

Axial Compressive Behavior of Cross-Shaped CFST Stub Columns with Steel Bar Truss Stiffening.

Concrete-filled steel tube (CFST) columns have been widely used in residential buildings due to their high bearing capacity, good ductility, and reliable seismic performance. However, conventional circular, square, or rectangular CFST columns may protrude from the adjacent walls, resulting in inconvenience in terms of the arrangement of furniture in a room. In order to solve the problem, special-shaped CFST columns, such as cross-shaped, L-shaped, and T-shaped columns, have been suggested and adopted in engineering practice. These special-shaped CFST columns have limbs with the same width as the adjacent walls. However, compared with conventional CFST columns, the special-shaped steel tube provides weaker confinement to the infilled concrete under axial compressive load, especially at concave corners. The parting at concave corners is the key factor affecting the bearing capacity and ductility of the members. Therefore, a cross-shaped CFST column with steel bar truss stiffening is suggested. In this paper, 12 cross-shaped CFST stub columns were designed and tested under axial compression loading. The effects of steel bar truss node spacing and column-steel ratio on the failure mode, bearing capacity, and ductility were discussed in detail. The results indicate that the columns with steel bar truss stiffening can change the final deformation mode of the steel plate from single-wave buckling to multiple-wave buckling, and the failure modes of columns also subsequently change from single-section concrete crushing failure to multiple-section concrete crushing failure. The steel bar truss stiffening shows no obvious effect on the axial bearing capacity of the member but significantly improves the ductility. The columns with a steel bar truss node spacing of 140 mm can only increase the bearing capacity by 6.8% while nearly doubling the ductility coefficient from 2.31 to 4.40. The experimental results are compared with those of six design codes worldwide. The results show that the Eurocode 4 (2004) and the Chinese code CECS159-2018 can be safely used to predict the axial bearing capacity of cross-shaped CFST stub columns with steel bar truss stiffening.

Artificial neural network assisted bearing capacity and confining pressure prediction for rectangular concrete-filled steel tube (CFT)

The evaluation of the confining pressure between steel and concrete is a complex yet important issue for concrete-filled steel tube (CFT) columns. The previous study is mainly based on the finite element model (FEM) because the confining pressure cannot be measured directly in the tests. Besides, the proposed approaches to calculating the bearing capacity of CFT columns and confining pressure varied significantly as a result of the different simplification. The neural network can simulate the complex problems quickly and accurately based on the proper training and validation. This study investigates the bearing capacity of CFT columns and the confining pressure of concrete based on a newly established experimental database and the artificial neural network (ANN) models. The results can be obtained from five primary parameters of rectangular CFT columns using the ANN models. The obtained bearing capacity and confining pressure from ANN models is compared and validated with that presented in the previous literatures. The parametric studies are performed to evaluate the bearing capacity and confining pressure with the variations of the dimensions and the material strength. The result could contribute to evaluate the interaction between the steel tube and infill concrete and its effect on the bearing capacity of rectangular CFT column.

Hysteresis performance of section steel reinforced concrete-filled rectangular stainless steel tubular columns

Due to high bearing capacity and excellent seismic performance, the section steel reinforced concrete-filled rectangular steel tube has been widely applied in engineering. The application of stainless steel for outer steel tube can improve the durability performance of members. In this study, ABAQUS software was used to perform nonlinear seismic analysis on section steel reinforced concrete-filled rectangular stainless steel tubular columns, and the numerical model was validated against experimental data. The mechanism and parametric study of the hysteresis performance were conducted. These parameters mainly include: concrete strength, strength of stainless steel, strength of section steel, steel ratio of stainless steel tube, steel ratio of section steel, axial compression ratio, and slenderness ratio. The analysis results showed that: increasing the strength of stainless steel tube and concrete can significantly improve the bearing capacity of members, but the strength of section steel has little effect; the bearing capacity decreases with the increasing slenderness ratio; when the axial compression ratio is smaller than 0.4, increasing axial compression can improve the bearing capacity. The strength of each material shows little influence on the shape of load–displacement skeleton; the axial compression ratio presents high effect on the shape of descending stage of curve, and the larger the axial compression ratio, the steeper the descending stage; the slenderness shows high influence on curve shape, and the larger the slenderness, the less the elastic stiffness, and the more gentle the descending stage of curve. Finally, a simplified load–displacement hysteretic model was proposed, which exhibits good agreement with finite element numerical analysis.

Numerical investigation and design recommendation for thin-walled octagonal concrete filled steel tubes

With the growing applications of high strength steel in modern constructions, thin-walled steel tubes attract many attentions. Local buckling failure is the most typical failure mode in a thin-walled steel tube, and concrete filled steel tubes (CFSTs) are excellent composite members in which the occurrence of local buckling is mitigated by the infilled concrete. This paper aims to investigate the compressive strength of concrete filled thin-walled octagonal steel tubes, which have the advantages of confinement efficiency over rectangular CFSTs, and constructability over circular CFSTs. Finite element models are established and validated by the experimental results from existing literatures. A parametric study is subsequently conducted and the result shows that the buckling strength of octagonal tube is significantly improved by the infilled concrete and the enhancement percentage increases with the increase of section slenderness coefficient, and the local buckling behaviour has a minor influence to the effect of confinement. In addition, a database consisting of 37 experimental specimens and 37 FE models are developed and used to assess the design methods in current codes of practice. It is found that the design methods from ANSI/AISC 360-16 and GB 50936-2014 underestimate the load capacity of octagonal CFSTs while the design formula from EN 1994-1-1 which considers the effect of confinement overestimates the capacity. Finally, a design model considering the buckling strength of octagonal steel tube, the influence of infilled concrete core and the effect of confinement is proposed and this model can well capture the compressive strength of octagonal CFSTs.

Feasibility of Stress Wave-Based Debond Defect Detection for RCFSTs Considering the Influence of Randomly Distributed Circular Aggregates with Mesoscale Homogenization Methodology.

In order to efficiently investigate the effect of the mesoscale heterogeneity of a concrete core and the randomness of circular coarse aggregate distribution on the stress wave propagation procedure and the response of PZT sensors in traditional coupling mesoscale finite element models (CMFEMs), firstly, a mesoscale homogenization approach is introduced to establish coupling homogenization finite element models (CHFEMs) with circular coarse aggregates. CHFEMs of rectangular concrete-filled steel tube (RCFST) members include a surface-mounted piezoelectric lead zirconate titanate (PZT) actuator, PZT sensors at different measurement distances, a concrete core with mesoscale homogeneity. Secondly, the computation efficiency and accuracy of the proposed CHFEMs and the size effect of representative area elements (RAEs) on the stress wave field simulation results are investigated. The stress wave field simulation results indicate that the size of an RAE limitedly affects the stress wave fields. Thirdly, the responses of PZT sensors at different measurement distances of the CHFEMs under both sinusoidal and modulated signals are studied and compared with those of the corresponding CMFEMs. Finally, the effect of the mesoscale heterogeneity of a concrete core and the randomness of circular coarse aggregate distribution on the responses of PZT sensors in the time domain of the CHFEMs with and without debond defects is further investigated. The results show that the mesoscale heterogeneity of a concrete core and randomness of circular coarse aggregate distribution only have a certain influence on the response of PZT sensors that are close to the PZT actuator. Instead, the interface debond defects dominantly affect the response of each PZT sensor regardless of the measurement distance. This finding supports the feasibility of stress wave-based debond detection for RCFSTs where the concrete core is a heterogeneous material.

Seismic behavior of composite shear walls with boundary columns using rectangular and U-shaped concrete-filled steel tubes

Steel-concrete composite shear walls with boundary concrete-filled steel tube (CFST) columns have drawn the attention of researchers due to their good seismic performance such as high strength, ductility and energy dissipation capacity. The reported study proposes a novel composite shear wall design that aims to complete the majority of the work in the factory, leaving only concrete pouring, steel welding, and formwork removal to be done on-site, so that the constructional time and cost can be saved. Two types of CFSTs were used for the boundary columns of composite shear walls, one is conventional rectangular CFSTs, another one is U-shaped CFSTs. In order to study the seismic behavior of the novel composite shear walls, cyclic loading tests were conducted on four full-scale specimens. The bearing capacity, energy dissipation, skeleton curves, hysteretic loops, deformation capacity and failure mode are analyzed. The experimental results have demonstrated that the bearing capacity and stiffness of shear walls with U-shaped CFST columns are slightly higher than that of shear walls with rectangular CFST columns, whereas the latter has better deformation capacity. In addition, the reported study investigated the methods of calculating the shear capacity of the novel composite shear walls and provided recommendations on the calculation method to be used for design and structural analysis.

Numerical analysis of novel concrete filled K joints with different brace sections

All over the world, truss structures are commonly used in long-span bridge. Since joint is the weakest part of this structure, this paper proposes a novel type in which different brace members are used in the concrete filled box section K-joints. This novel type is designed as rectangular compression brace with the brace width to chord ratio β<0.8 and chord welded tension brace (β=1). That configuration allows to reduce the gap, thus eliminate the secondary moment. Moreover, load transfer and failure modes are not similar to the ordinary. Numerical simulation is the used method of investigation in which validation is carried out through thirty-four models composed of RHS K gap Joint – CFST T Joint – CFST Y Joint – RHS T Integral Joint – CFST K gap Joint. Difference between experiments and Finite Element models are less than 20% thus acceptable. Through the validated numerical simulation model, analysis of suitable boundary conditions and variation of initial stiffness, ultimate strength are presented in accordance with the novel joint parameters. Comparison of the novel joint type to the rectangular hollow section (RHS) and rectangular concrete filled steel tube (RCFST) is carried out in terms of initial stiffness and ultimate strength. Finally, optimization of the novel type of joint is proposed for engineering practice so as to have overview of the strength in practical aspect. Results have shown that most of the proposed types of boundary condition procure the joint deformation while subjected in compression and tension load. Tension brace failure is the typical failure mode of the novel joint in which the chord width, the parameter β are fundamental and directly proportional to the joint initial stiffness and ultimate strength. For β=0.8 and the chord width between 500 and 1000 mm, the initial stiffness can vary from [9944.92kN/mm−19887.31kN/mm]; ultimate strength varies from [29551.76kN−117916.20kN]. Moreover, the novel joint type is stronger than the RHS and the RCFST in both initial stiffness and ultimate strength. That difference is about 3–6% in the initial stiffness and around 10% for the ultimate strength. These demonstrate that the novel type of joint is acceptable in engineering truss bridge with proposition of joint optimization.

Study on fatigue performance of double cover plate through-core bolted joint of rectangular concrete-filled steel tube bundle wind turbine towers

Rectangular steel tube concrete bundle wind turbine towers is a new type of tower structure with high prefabrication, convenient installation, convenient transportation and good integrity.Based on the uniaxial tensile test and fatigue loading test of the double cover plate through-core bolted joint, the stress condition of the key connection parts of the structure is analyzed, the S-N curve of the joint under parameter expansion is summarized, and the Weibull distribution test is carried out for the fatigue life of the concrete inside the joint. Finally, the fatigue damage of the joint are analyzed. The results show that the failure mode of the double cover plate through-core bolted joint under fatigue loading is bolt shear; Through S-N curve fitting, the S-N curve expression and fatigue limit value of different types of joints are obtained. It is found that under the shear failure mode, the fatigue limit of different types of joints is higher than the limit value specified in GB50017–2017, EC3 and AISC-360; The fatigue life of concrete at different loading factors conforms to the two parameter Weibull distribution.The damage analysis of the joint through the high cycle fatigue damage model shows that the trend of a sharp increase in the damage rate of the joints in the late loading period can better explain the phenomenon of instantaneous fracture. The research results of this paper can provide reference for the analysis, research and application of wind turbine towers.

New cross-sectional slenderness limits for stiffened rectangular concrete-filled steel tubes

Rectangular concrete-filled steel tubes (RCFST) are efficient structural members that exhibit good strength, stiffness, and ductility. The mechanical properties of RCFST can be further enhanced by the addition of internal stiffeners which delay local buckling of the tube, improve load transfer, and can confine the concrete. Numerous experimental and analytical studies have been performed to characterize the behavior of stiffened RCFST. However, most studies focus on a single type of stiffener. Also, most major design standards do not provide guidance on how to account for the beneficial effects of the stiffeners. The objective of this work was to develop strength design recommendations for short stiffened RCFST columns and beam-columns that are applicable to a variety of stiffener types. A database of previously published experimental tests on stiffened RCFST members was compiled and the experimental results were compared to American, European, and Chinese design standards. Each of these standards employs the plastic stress distribution method for filled composite members that are not susceptible to local buckling as identified using cross-sectional slenderness ratio limits. The results indicate that neglecting the stiffeners is unnecessarily conservative. New cross-sectional slenderness ratio limits are proposed for stiffened RCFST, allowing a greater range of members to be assessed using the plastic stress distribution method. These results provide strong justification for changes to design standards to recognize the benefits of highly efficient stiffened RCFST members.

Axial compression capacity of rectangular CFST columns under random pitting corrosion

To study the effect of random pitting corrosion, this paper presents the change law of the bearing capacity of the rectangular concrete-filled steel tube (CFST) column under the action of axial force with the mass loss ratio of the outer steel tube. Existing experimental data, including the axial ultimate bearing capacity and load displacement curves of the uncorroded and corroded rectangular CFST columns, are collected to verify the rationality of the finite element model. ANSYS software was used to establish a model that can simulate the pitting corrosion, and the results of the finite element analysis were compared with the experiment, which illustrates that the established model can accurately predict the actual load changes. For further research, the corroded thickness tc/t, which include constant tc/t and random tc/t, and other influencing factors such as corrosion pit diameter Dc, steel tube wall thickness t, material strength, shape and section size of specimens are discussed, the results reveal the effect of these parameters on the decline of axial compressive load capacity. Meanwhile, to visually reflect this effect, this study introduces the concept of bearing capacity reduction coefficient Rcs and Rc to express the ratio of bearing capacity of specimens before and after corrosion. The extensive stochastic finite element analysis was carried out, and on this basis, a suitable calculation formula for the reduction factor is proposed to predict the axial compression bearing capacity of the corroded rectangular CFST column by measuring the value of local mass loss ratio of the outer steel.

Behaviour of Cold Formed Steel Rectangular Hollow Section Filled with Concrete under Cyclic Lateral Loadwith Constant High Axial Compression

This paper presents an experimental investigation of Rectangular Concrete Filled Steel Tube (RCFST) sections subjected to cyclic lateral load combined with constant high axial compression. A total of eight RCFST columns with different combinations of width to thickness ratio, slenderness ratio and axial compression ratio were tested. The constant axial compressive load on the test columns was 60% and 80% of its axial load carrying capacity. Failure modes such as local buckling and rupture of steel tube were observed from experiments during strength degradation after reaching maximum lateral load. From the experimental results, envelope curves were obtained by joining the locus of extremities of the load-displacement hysteresis loops. Seismic parameters such as strength, stiffness, ductility, and energy dissipation capacity were discussed based on experimental results.

Machine learning (ML) based models for predicting the ultimate strength of rectangular concrete-filled steel tube (CFST) columns under eccentric loading

Concrete-filled steel tubes (CFSTs) are popularly used in structural applications. The accurate prediction of their ultimate strength is a key for the safety of the structure. Extensive studies have been conducted on the strength prediction of CFSTs under concentric loading. However, in real situation CFSTs are usually subjected to eccentric loading. The combined compression and bending will result in more complex failure mechanisms at the ultimate strength. The accuracy of methods in design codes is usually limited due to their simplicity. In this study, three machine learning (ML) methods, namely, Support Vector Regression (SVR), Random Forest Regression (RFR), and Neural Networks (NN), are adopted to develop models to predict the ultimate strength of CFSTs under eccentric loading. A database consisting of information of 403 experimental tests from literature is created and statistically analyzed. The database was then split to a training set which was used to optimize and train the ML models, and a test set which was used to evaluate performance of trained ML models. Compared with the methods in two typical design codes, the ML models achieved notable improvement in prediction accuracy. The parametric study revealed that the trained ML models could generally capture the effect of each primary input feature, which was verified by the relevant experimental test results.

Optimized data-driven machine learning models for axial strength prediction of rectangular CFST columns

The nonlinear material interaction in concrete-filled steel tube (CFST) significantly contributes to its excellent axial compression behaviour, which in the meantime results in highly complex relationships between various specimen parameters and the column strength. Machine learning (ML) technique, being excellent in capturing complicated data mapping, is therefore applied in this study to predict the axial compression strength of rectangular CFST columns. A comprehensive test database containing 1,641 rectangular CFST samples is established. The key input parameters for ML models are identified through both correlation analysis and mechanical principles, highlighting the effects of sectional configuration and column slenderness. Strength prediction models are established applying five mainstream ML methods including the back-propagation neural network, the radial basis function neural network, the adaptive neuro-fuzzy inference system, the Gaussian process regression and the M5′ model tree. The prediction reasonableness and stability of established models are comprehensively evaluated and compared. Outcomes reveal that the established ML models exhibit higher prediction accuracies and wider applicable ranges than current design standards, whilst the prediction stability of ML models is highly affected by the quantity of the available data set. Finally, ML models are optimized by dividing the training database as per mechanical principles, and a prediction-error-ratio amending method based on ML algorithm is also proposed to further improve the model accuracy.

Novel ensemble approach to predict the ultimate axial load of CFST columns with different cross-sections

Concrete-filled steel tubes (CFST) have been widely used in construction due to their numerous benefits over traditional structural components made of steel or reinforced concrete. The ultimate axial load (UAL) of the CFST column, regarded as the most critical mechanical property, is determined by several constitutive elements, including the material’s mechanical properties and tube cross-sections. Due to the complexity introduced by the interaction of steel tubes and the filling concrete, it is challenging to conduct an analysis of CFST columns using design codes and empirical formulas. As a consequence of this, the principal objective of this study is to use machine learning (ML) methods to predict the UAL of CFST columns. This study mentions six ML models, divided into two groups: the single (or individual) and ensemble ML models. The considered models include Decision Tree, Gradient Boosting, Light Gradient Boosting, Adaptive-Gradient Boosting, Voting Ensemble, and Stacking Ensemble. A database containing 349 test results defining the UAL of rectangular and elliptical CFST columns is compiled to construct the ML models. For model assessment purposes, common quality evaluation criteria, a 10-fold cross-validation approach, and Monte Carlo simulations are utilized. The observed results demonstrate that ensemble ML models outperform single ML models in terms of performance and stability, with the Adaptive-Gradient Boosting model being the best predictive model. In the last step, a partial dependence plot (PDP) analysis is carried out to assess the impact of each input variable and the correlation between variables on the UAL of the CFST columns.

Cyclic Performance of Prefabricated Shear Wall Connected to Columns by Rectangular Concrete-Filled Steel Tube Keys

A prefabricated frame–shear wall structure is a major structure in an assembled building. To find a method of connecting a shear wall and columns that can both reduce the amount of wet work required and maintain adequate stiffness, we propose connecting the shear wall to the vertical frame using rectangular concrete-filled steel tube keys (RCFSTKs). Static tests of a cast-in-place frame–shear wall structure without keys and a prefabricated structure with RCFSTKs were conducted to compare their seismic performance. The feasibility of the new method was analyzed. Finite element models were then set up to determine if plain concrete blocks between RCFSTKs could be removed and to identify the influence of different parameters on the cyclic performance of the assembled structure. The results show that the use of RCFSTK is practical. Compared with a dimensionally similar cast-in-place shear wall–column construction, a prefabricated shear wall connected to columns by RCFSTKs has a fuller hysteresis curve, better ductility, slightly higher energy dissipation, and slightly slower degradation of stiffness and bearing capacity. The removal of inter-key concrete blocks significantly reduces bearing capacity and initial stiffness. The cyclic performance of the assembled structure is primarily influenced by the number of RCFSTKs, the thickness of the steel tube key wall, and the axial compression ratio, with less effect from key unit height, steel strength, and concrete grade in RCFSTKs.

Seismic behavior of concrete-filled wide rectangular steel tubular (CFWRST) stub columns

The concrete-filled wide rectangular steel tube (CFWRST) columns, with the aspect ratio of cross-section greater than 2.0, have attracted much attention in practice. Existing studies on the concrete-filled rectangular steel tube (CFRST) members have shown that the aspect ratio of cross-section is a main parameter controlling their behaviors under various loads, which means that the behaviors of CFWRST columns may be different from the common CFST members (the aspect ratio of cross-section not greater than 2.0) considered in existing studies. The behaviors of CFWRST columns under the combined constant axial compression and cyclic lateral loadings were investigated in this paper. Six full-scale columns were tested, with which the failure modes, hysteresis curve, skeleton curve, ductility, lateral stiffness and energy dissipation capacity of the specimens were analyzed and discussed in detail. For all tested specimens, the ductility indexes are between 2.028 and 3.433, and the equivalent damping ratios usually exceed 0.3 at the last of loading cycle. The finite element model was also developed and validated using the test results, and the parametric analyses were carried out with the influences of main parameters discussed. Furthermore, comparisons between the predictions of the solution of EC4 for CFST beam-columns and test results show that the solution of EC4 can provide safe predictions for the CFWRST columns under constant axial compression and cyclic lateral loading.

Analysis of Buckling Deformation for the Side Plate of Rectangular CSFT Column Based on Plate Theory with Bi-Axial Loads

In this paper, under the condition of bidirectional stress, the buckling deformation of the side plate in a rectangular concrete-filled steel tube (CFST) column has been studied in detail. We have conducted a theocratical analysis, an experimental validation and a finite element simulation to investigate the influences of the height-width ratios and Nominal Poisson’s ratios on the buckling form of the side plate, and we also try to explain the change of buckling form between unidirectional and bidirectional stress, both of them can provide a good reference and basis for design and application of the CFST column. The specific work can be summarized as follows: Firstly, a theoretical analysis has been conducted to study the buckling coefficient solution method of the thin plate under the conditions of axial compress and transverse tension. Then, under the conditions of the unidirectional and the bidirectional stress, a comparative study is carried out to investigate the changing relationship of the buckling coefficient (k) of the side plate; the results indicate that the buckling characteristic is changed due to the bidirectional stress, meanwhile, the buckling coefficient and the number of buckling half-wave will increase. Furthermore, the existing outcomes and the numerical simulations are adopted to study the relevance between the number of the elastic buckling half-wave in the side plate and the corresponding height-width ratio of the component; the results indicate that the former is larger than the latter. Finally, based on the obtained, the buckling relationship curve, the conclusion can be drawn as follows: when the bidirectional stress has been applied to the side plate, there is an equal interval between the different buckling half-waves; meanwhile, the interval shows a quadratic function reduce trend with the increase of nominal Poisson’s ratio.

Hybrid intelligence approach for performance estimation of rectangular CFST columns under different loading conditions

The ultimate bearing capacity (UBC) is a crucial design parameter of compressed rectangular concrete-filled steel tube (RCFST) members. Past extensive experimental studies have demonstrated that the UBC of RCFST has a highly nonlinear relationship with its constitutes due to the sophisticated interaction mechanism between steel tube and concrete core. 1) Although massive experimental data are available in literatures, conventional methods still struggle in accurately modeling and predicting the UBC of RCFST based on existing dataset. 2) RCFST columns with various loading conditions are treated differently in an analytical method, which does not satisfy the consistency of modern design. In this study, a hybrid data-driven model is developed to predict the UBC of both axially and eccentrically loaded RCFST using multilayer perceptron (MLP) neural network integrated with beetle antennae search (BAS) algorithm, abbreviated as BAS-MLP. MLP is developed using existing experimental data, while BAS is introduced to optimize the initial connection weights and biases of MLP to further enhance its nonlinear mapping capacity. In the proposed BAS-MLP model, input variables include geometrical and material properties of RCFST columns with different loads, and the model output is UBC. The proposed model is evaluated using additional testing datasets, and the prediction accuracy of BAS-MLP is verified by comparing obtained results with those from baseline models using conventional modeling methods. A global sensitivity analysis is further conducted to measure the contribution of each input variable affecting the UBC of RCFST. The results show that the adaptive BAS-MLP model performs better than several baseline models and conventional methods (analytical methods and numerical simulation), and geometrical properties of RCFST columns are highly correlated with UBC. As a result, it can be regarded as an effective tool to predict the mechanical behavior of RCFST columns and other composite members, thereby reducing the cost and time in material planning and design.