Sürdürülebilir Kalkınma Amaçları

• 2
• 1

• MakaleYalnız künye2026Scopus SJRQ1/r/munzur/647
  Flexural response prediction of C-shaped concrete-filled composite plate shear walls (C-PSW/CF) using response surface methodology

  Polat, Erkan

  Concrete-filled composite plate shear walls (C-PSW/CF) are an emerging prefabricated structural system that combines high strength, stiffness, and constructability for use in high-rise and seismic applications. Conventional plastic-theory-based flexural models originally developed for planar walls are intended only to estimate peak flexural strength; however, they cannot adequately capture several mechanisms governing the behavior of C-shaped sections, including concrete confinement, plate local buckling, von Mises stress interaction, and shear-lag effects. Moreover, no analytical formulations are currently available for predicting the full monotonic flexural response of such walls. This study develops a surrogate-based modeling framework that integrates high-fidelity finite element (FE) simulations with Response Surface Methodology (RSM) to predict the monotonic flexural response and yield moment capacity of C-shaped C-PSW/CF walls. A validated three-dimensional FE model (FEM) was used to simulate 77 wall configurations generated through a Central Composite Design (CCD) by varying six geometric and material parameters. Second-order RSM equations were derived to estimate moment capacity across drift ratios up to ±4% in both loading directions and to predict yield moments corresponding to the onset of web and flange plate yielding. The developed models show high predictive accuracy (R² > 0.95) and successfully capture nonlinear parameter interactions among section geometry, reinforcement ratio, and steel yield strength. Compared with nonlinear FE simulations, the RSM equations enable rapid prediction of wall capacity using explicit closed-form expressions, providing a computationally efficient tool for performance-based design and parametric evaluation of prefabricated composite wall systems.

  AtıfScopus1·WoS1·OpenAlex1

  İncele
• MakaleYalnız künye2024Scopus SJRQ1/r/munzur/2134
  Assessment of short and long-term pozzolanic activity of natural pozzolans using machine learning approaches

  Khatti, Jitendra; Polat, Berivan Yılmazer

  This investigation introduces the optimal performance models for predicting the compressive strength (CS) and pozzolanic activity index (PAI) by comparing the machine learning models. The machine learning models, i.e., multilinear regression (MLR), support vector machine (SVM), gaussian process regression (GPR), decision tree (DT), random forest (RF), and gene expression programming (GEP) have been trained (TRN) and tested (TST) by 28 and 7 data points. For the first time, the SiO2, Al2O3, Fe2O3, SiO2 +Al2O3 +Fe2O3, reactive SiO2, Blaine specific surface area, and specific gravity have been used as input variables to compute the CS, and 28 days PAI (28PAI), and 90 days PAI (90PAI) of the natural pozzolans. The multicollinearity analysis showed the SiO2, Al2O3, Fe2O3, SiO2 +Al2O3 +Fe2O3, reactive SiO2, and specific gravity have problematic multicollinearity (variance inflation factor – VIF > 10). Therefore, the root mean square error (RMSE), mean absolute error (MAE), correlation coefficient (R), performance index (PI), and variance accounted for (VAF) metrics have been implemented to evaluate the model's performance and multicollinearity impact. From the comparison of models, it has been recorded that model GPR outperformed the MLR, SVM, DT, RF, and GEP models in predicting CS (PI = 1.29, VAF = 71.31, R = 0.8473, MAE = 0.9390 MPa), 28PAI (PI = 1.87, VAF = 94.88, R = 0.9744, MAE = 0.7295 %), and 90PAI (PI = 1.72, VAF = 88.11, R = 0.9393, MAE = 1.2444 %) in the TST phase, close to ideal values. The score, generalizability. Wilcoxon test, uncertainty analysis, Anderson-daring test, and accuracy metrics have confirmed the superiority of GPR models in predicting CS, 28PAI, and 90PAI of natural pozzolans.

  • 
  AtıfScopus28·WoS25·OpenAlex28

  İncele
• MakaleYalnız künye2024Scopus SJRQ1/r/munzur/1909
  Predicting and optimizing maximum flexural strength of planar composite plate shear walls – concrete filled with the application of LR and RSM

  Polat, Erkan; Polat, Elifcan Göçmen

  Traditional theoretical models, while conservative, overlook the complex, nonlinear interactions between material and geometric parameters that influence wall flexural strength. Additionally, the extent to which these models maintain their conservative assumptions across varying parameters remains uncertain. This study aims to develop advanced predictive models for the flexural strength of planar composite plate shear walls—concrete filled (C-PSW/CF) by employing Linear Regression (LR) and Response Surface Methodology (RSM). Central Composite Design (CCD) was utilized to design numerical experiments to analyze the effects of varying parameters, including steel plate yield strength, concrete compressive strength, wall depth, depth-to-thickness ratio, and reinforcement ratio. A validated complex finite element modeling (FEM) procedure was used to analyze numerical experiments involving varying parameters of the selected variables. Both LR and RSM were applied to the generated dataset to formulate predictive equations for the maximum moment capacity of walls. Interaction plots were developed to reveal the nonlinear relationships between the variables and identify the most effective parameters influencing the wall strength. Comparisons between theoretical predictions, RSM, and LR models revealed that RSM provided the most accurate predictions, closely aligning with FEM results and effectively capturing the nonlinear interactions between wall parameters. The findings of this paper contribute to the development of more practical, accurate and optimized design methods for C-PSW/CF, ultimately improving both structural safety and efficiency.

  AtıfScopus10·WoS8·OpenAlex10

  İncele
• MakaleYalnız künye2020Scopus SJRQ1/r/munzur/3826
  Impact of mechanical properties of historical masonry bridges on fundamental vibration frequency

  Onat, Onur

  The fundamental frequency of a historical monument like a masonry bridge or any other structure provides a better and detailed seismic assessment of its demand to provide protection against unexpected seismic ground motion. On this basis, this paper aims to develop an empirical formulation to fit the ambient vibration frequency of historical masonry bridges using a nonlinear regression model. For this purpose, data series were collected from the literature especially focused on both ambient vibration measurement and finite element models of historical masonry bridges modeled on a full scale to get the first global natural frequency. The first approach is to forecast the natural vibration frequency by using only physical characteristics of the historical masonry bridges obtained from the literature-based experimental ambient vibration frequency data. The second approach consists of forecasting the first natural vibration frequency considering the unit weight and elasticity modulus of the dominant construction material based on the homogenization approach. In addition to mechanical characteristics, physical properties of historical masonry bridges were used, such as the length, height, width, and Main Arch Span (MAS) length of the masonry bridge, to predict the first natural vibration frequency. Among the proposed equations, a maximum accuracy of 58% was reached with the literature-based experimental approach. Moreover, an empirical formulation with 81% accuracy was proposed by using both physical characteristics and mechanical properties of the bridges on the basis of finite element model results. Also, this study highlights that this accuracy decreases to 35% if the modulus of elasticity and unit weight are ignored. In addition, the developed formulations are compared with other empirical formulations using the Mean Square Error (MSE), as a measure of the average of the squares of the errors. Consequently, the smallest MSE value of 1.95 was obtained with the proposed equation, whose accuracy was 81% in this study.

  • 
  • 
  AtıfScopus17·WoS19·OpenAlex21

  İncele