• Steel and Composite Structures
• /
• 제52권2호
• /
• pp.145-163
• /
• 2024

This paper presents six novel hybrid machine learning (ML) models that combine support vector machines (SVM), Decision Tree (DT), Random Forest (RF), Gradient Boosting (GB), extreme gradient boosting (XGB), and categorical gradient boosting (CGB) with the Harris Hawks Optimization (HHO) algorithm. These models, namely HHO-SVM, HHO-DT, HHO-RF, HHO-GB, HHO-XGB, and HHO-CGB, are designed to predict the ultimate strength of both rectangular and circular reinforced concrete (RC) columns. The prediction models are established using a comprehensive database consisting of 325 experimental data for rectangular columns and 172 experimental data for circular columns. The ML model hyperparameters are optimized through a combination of cross-validation technique and the HHO. The performance of the hybrid ML models is evaluated and compared using various metrics, ultimately identifying the HHO-CGB model as the top-performing model for predicting the ultimate shear strength of both rectangular and circular RC columns. The mean R-value and mean a20-index are relatively high, reaching 0.991 and 0.959, respectively, while the mean absolute error and root mean square error are low (10.302 kN and 27.954 kN, respectively). Another comparison is conducted with four existing formulas to further validate the efficiency of the proposed HHO-CGB model. The Shapely Additive Explanations method is applied to analyze the contribution of each variable to the output within the HHO-CGB model, providing insights into the local and global influence of variables. The analysis reveals that the depth of the column, length of the column, and axial loading exert the most significant influence on the ultimate shear strength of RC columns. A user-friendly graphical interface tool is then developed based on the HHO-CGB to facilitate practical and cost-effective usage.

• 한국재료학회지
• /
• 제20권11호
• /
• pp.623-627
• /
• 2010

Semiconducting metal oxides have been frequently used as gas sensing materials. While zinc oxide is a popular material for such applications, structures such as nanowires, nanorods and nanotubes, due to their large surface area, are natural candidates for use as gas sensors of higher sensitivity. The compound ZnO has been studied, due to its chemical and thermal stability, for use as an n-type semiconducting gas sensor. ZnO has a large exciton binding energy and a large bandgap energy at room temperature. Also, ZnO is sensitive to toxic and combustible gases. The NO gas properties of zinc oxide-single wall carbon nanotube (ZnO-SWCNT) composites were investigated. Fabrication includes the deposition of porous SWCNTs on thermally oxidized $SiO_2$ substrates followed by sputter deposition of Zn and thermal oxidation at $400^{\circ}C$ in oxygen. The Zn films were controlled to 50 nm thicknesses. The effects of microstructure and gas sensing properties were studied for process optimization through comparison of ZnO-SWCNT composites with ZnO film. The basic sensor response behavior to 10 ppm NO gas were checked at different operation temperatures in the range of $150-300^{\circ}C$. The highest sensor responses were observed at $300^{\circ}C$ in ZnO film and $250^{\circ}C$ in ZnO-SWCNT composites. The ZnO-SWCNT composite sensor showed a sensor response (~1300%) five times higher than that of pure ZnO thin film sensors at an operation temperature of $250^{\circ}C$.

• Advances in nano research
• /
• 제9권4호
• /
• pp.237-250
• /
• 2020

An accurate plate theory for assessing sandwich structures is of interest in order to provide precise results. Hence, this paper develops Layer-Wise (LW) theory for reaching precise results in terms of buckling and vibration behavior of Functionally Graded Carbon Nanotube-Reinforced Composite (FG-CNTRC) annular nanoplates. Furthermore, for simulating the structure much more realistic, its edges are elastically restrained against in-plane and transverse displacement. The nano structure is integrated with piezoelectric layers. Four distributions of Single-Walled Carbon Nanotubes (SWCNTs) along the thickness direction of the core layer are investigated. The Differential Quadrature Method (DQM) is utilized to solve the motion equations of nano structure subjected to the electric field. The influence of various parameters is depicted on both critical buckling load and frequency of the structure. The accuracy of solution procedure is demonstrated by comparing results with classical edge conditions. The results ascertain that the effects of different distributions of CNTs and their volume fraction are significant on the behavior of the system. Furthermore, the amount of in-plane and transverse spring coefficients plays an important role in the buckling and vibration behavior of the nano-structure and optimization of nano-structure design.

• Steel and Composite Structures
• /
• 제20권3호
• /
• pp.545-569
• /
• 2016

Stiffeners have widely been used in lateral load resisting systems to improve the buckling stability of shear panels in steel frames. However, due to major differences between plate girders and steel plate shear walls (SPSWs), use of plate girder equations often leads to uneconomical and, in some cases, incorrect design of stiffeners. Hence, this paper uses finite element analysis (FEA) to describe the effect of the rigidity and arrangement of stiffeners on the buckling behavior of plates. The procedures consider transverse and/or longitudinal stiffeners in various practical configurations. Subsequently, curves and formulas for the design of stiffeners are presented. In addition, the influence of stiffeners on the inward forces subjected to the boundary elements and the tension field angle is investigated as well. The results indicate that the effective application of stiffeners in SPSW systems not only improves the structural behavior, such as stiffness, overall strength and energy absorption, but also leads to a reduction of the forces that are exerted on the boundary elements.

• Steel and Composite Structures
• /
• 제17권2호
• /
• pp.173-184
• /
• 2014

This study aimed to theoretical calculate the thermal residual stress in continuous SiC fiber reinforced titanium matrix composites. The analytical solution of residual stress field distribution was obtained by using coaxial cylinder model, and the numerical solution was obtained by using finite element model (FEM). Both of the above models were compared and the thermal residual stress was analyzed in the axial, hoop, radial direction. The results indicated that both the two models were feasible to theoretical calculate the thermal residual stress in continuous SiC fiber reinforced titanium matrix composites, because the deviations between the theoretical calculation results and the test results were less than 8%. In the titanium matrix composites, along with the increment of the SiC fiber volume fraction, the longitudinal property was improved, while the equivalent residual stress was not significantly changed, keeping the intensity around 600 MPa. There was a pronounced reduction of the radial residual stress in the titanium matrix composites when there was carbon coating on the surface of the SiC fiber, because carbon coating could effectively reduce the coefficient of thermal expansion mismatch between the fiber and the titanium matrix, meanwhile, the consumption of carbon coating could protect SiC fibers effectively, so as to ensure the high-performance of the composites. The support of design and optimization of composites was provided though theoretical calculation and analysis of residual stress.

• Steel and Composite Structures
• /
• 제30권3호
• /
• pp.271-280
• /
• 2019

In this paper, numerical and experimental assessments have been conducted in order to investigate the capability of using CFRP for the seismic capacity improvement and relocation of plastic hinge in reinforced concrete connections. Two scaled down exterior reinforced concrete beam to column connections have been used. These two connections from a strengthened moment frame have been tested under uniformly distributed load before and after optimization. The results of experimental tests have been used to verify the accuracy of numerical modeling using computational ABAQUS software. Application of FRP plate on the web of the beam in connections to improve its capacity is of interest in this paper. Several parametric studies were carried out for CFRP reinforced samples, with different lengths and thicknesses in order to relocate the plastic hinge away from the face of the column.

• Steel and Composite Structures
• /
• 제43권1호
• /
• pp.55-66
• /
• 2022

To develop high-efficiency lateral force resistance components for high-rise buildings, a novel energy dissipation shear wall with concrete-filled steel tubular (CFST) column elements was proposed. An energy dissipation shear wall specimen with CFST column elements (GZSW) and an ordinary reinforced concrete shear wall (SW) were constructed, and experimented by low-cycle reversed loading. The mechanical characteristics of these two specimens, including the bearing capacity, ductility, energy dissipation, and stiffness degradation process, were analyzed. The finite-element model of the GZSW was established by ABAQUS. Based on this finite-element model, the effect of the placement of steel-plate energy dissipation connectors on the seismic performance of the shear wall was analyzed, and optimization was performed. The experiment results prove that, the GZSW exhibited a superior seismic performance in terms of bearing capacity, ductility, energy dissipation, and stiffness degradation, in comparison with the SW. The results calculated by the ABAQUS finite-elements model of GZSW corresponded well with the results of experiment, and it proved the rationality of the established finite-elements model. In addition, the optimal placement of the steel-plate energy dissipation connectors was obtained by ABAQUS.

• Steel and Composite Structures
• /
• 제46권6호
• /
• pp.805-821
• /
• 2023

This study developed an evaluation system to explore the effect of the environmental temperature on the stress redistribution produced by cable stress relaxation of structural members in a steel cable-stayed bridge. The generalized Maxwell model is used to estimate stress relaxation at different temperatures. The environmental temperature is represented using the thermal coefficients and temperature loads. The fmincon optimization function is used to determine the set of stress relaxation parameters at different temperatures for all cables. The ABAQUS software is employed to investigate the stress redistribution of the steel cable-stayed bridge caused by the cable stress relaxation and the environmental temperature. All of these steps are set up as an evaluation system to save time and ensure the accuracy of the study results. The developed evaluation system is then employed to investigate the effect of environmental temperature and cable type on stress redistribution. These studies' findings show that as environmental temperatures increased up to 40 ℃, the redistribution rate increased by up to 34.9% in some girders. The results also show that the cable type with low relaxation rates should be used in high environmental temperature areas to minimize the effect of cable stress relaxation.

• Steel and Composite Structures
• /
• 제44권6호
• /
• pp.867-882
• /
• 2022

Predicting the compressive strength of concrete (CSoC) is of high significance in civil engineering. The CSoC is a highly dependent and non-linear parameter that requires powerful models for its simulation. In this work, two novel optimization techniques, namely evaporation rate-based water cycle algorithm (ER-WCA) and equilibrium optimizer (EO) are employed for optimally finding the parameters of a multi-layer perceptron (MLP) neural processor. The efficiency of these techniques is examined by comparing the results of the ensembles to a conventionally trained MLP. It was observed that the ER-WCA and EO optimizers can enhance the training accuracy of the MLP by 11.18 and 3.12% (in terms of reducing the root mean square error), respectively. Also, the correlation of the testing results climbed from 78.80% to 82.59 and 80.71%. From there, it can be deduced that both ER-WCA-MLP and EO-MLP can be promising alternatives to the traditional approaches. Moreover, although the ER-WCA enjoys a larger accuracy, the EO was more efficient in terms of complexity, and consequently, time-effectiveness.

• Steel and Composite Structures
• /
• 제44권6호
• /
• pp.753-768
• /
• 2022

In this study, a graphical automatic system is developed in order to investigate the stress redistribution of structural members in a steel cable-stayed bridge. The generalized Maxwell model is selected for stress relaxation estimation, and it is carefully verified and applied to all the cable members of a steel cable-stayed bridge to investigate its stress relaxation. A set of stress relaxation parameters in all cables is determined using the fmincon optimization function. The stress redistribution of the steel cable-stayed bridge is then analyzed using ABAQUS. To shorten the investigation time, all the aforementioned phases are built up to be an automatic system. The automatic system is then employed to investigate the effect of cable cross-section areas and girder spans on stress redistribution. The findings from these studies show that the initial tension in the cables of a steel cable-stayed bridge should be kept to less than 55% of the cable's ultimate strength to reduce the effect of cable stress relaxation. The cable space in a steel cable-stayed bridge should be limited to 15,000 mm to minimize the effect of cable stress relaxation. In comparison to other structural members of a steel cable-stayed bridge, the girders experience a significant stress redistribution.