Browsing by Author "Yildizel, Sadik Alper"

Now showing 1 - 6 of 6

Compressive Strength Optimization of Natural Zeolite-Based Geopolymers Via Taguchi Design, Grey Relational Analysis, and Genetic Algorithms
(IOP Publishing Ltd, 2025) Liban, Roble Ibrahim; Turk, Furkan; Keskin, Ulku Sultan; Yildizel, Sadik Alper
This study examines the optimization of natural zeolite-based ternary geopolymer mortars via an integrated Taguchi-Grey Relational Analysis-Genetic Algorithm (Taguchi-GRA-GA) framework to improve mechanical performance and durability. Natural zeolite (NZ) was partially substituted (up to 50 wt%) with fly ash (FA) and calcium hydroxide (CH) to enhance binder reactivity and matrix density. A Taguchi L9 orthogonal design was utilized to determine ideal activator values, subsequently employing GRA to amalgamate compressive strength findings at 7, 28, and 90 days into a singular performance index.The top-performing mixtures (F20C20, F25C25, and F30C30) were experimentally validated and utilized to develop regression-based predictive models for subsequent GA optimization. The genetic algorithm identified an optimal formulation (NZ = 214.6 g dm-3, FA = 116.4 g dm-3, CH = 116.4 g dm-3) that achieved a predicted compressive strength of 33.01 MPa, with experimental validation showing a deviation of less than 1.1%. This integrated method demonstrates that the combination of statistical design, data-driven modeling, and evolutionary optimization provides an efficient strategy for developing sustainable, high-performance binders. The resulting materials enhance strength and durability, allowing low-carbon, sustainable construction solutions aligned with global sustainability objectives.
Enhancing Mechanical Performance of Glass Fiber Reinforced Gypsum Composites With Carbon Black and Magnetite: an Integrated Optimization Approach
(Elsevier, 2025) Yildizel, Sadik Alper; Toktas, Abdurrahim; Keskin, Ulku Sultan
This study presents a comprehensive optimization methodology that integrates Taguchi Design of Experiments (Taguchi DoE), data augmentation, and the Cuckoo Search Algorithm (CSA) to improve the mechanical and electromagnetic characteristics of gypsum-based composites reinforced with carbon black, magnetite, and glass fiber. The effects of these additives on compressive and flexural strengths were evaluated using an L-16 orthogonal array, and optimal mixes were determined. The hybrid model attained a compressive strength of 32.23 MPa and a flexural strength of 1.41 MPa, demonstrating remarkable prediction accuracy (R-2 > 0.95). The integrated approach also allows cost-effective creation of multifunctional gypsum composites with improved mechanical and electromagnetic properties, in line with advanced construction material development.
Optimization of Concrete with Human Hair Using Experimental Study and Artificial Neural Network via Response Surface Methodology and Anova
(Nature Portfolio, 2025) Yildizel, Sadik Alper; Karalar, Memduh; Aksoylu, Ceyhun; Althaqafi, Essam; Beskopylny, Alexey N.; Stel'makh, Sergey A.; Ozkilic, Yasin Onuralp
The increasing demand for sustainable construction materials has prompted the investigation of non-biodegradable waste, such as human hair (HH), for concrete reinforcement. This study seeks to evaluate the impact of HH fiber on the fresh, physical, and mechanical characteristics of concrete. HH was incorporated in varying proportions (1-5% by weight of cement), along with modifications in cement content, to ascertain optimal performance conditions. An extensive experimental program was executed, succeeded by the utilization of Artificial Neural Networks (ANN) to formulate predictive models for compressive strength (CS), flexural strength (FS), and splitting tensile strength (STS). Furthermore, Response Surface Methodology (RSM) and Analysis of Variance (ANOVA) were utilized to identify statistically significant factors and optimize the mix design. The findings indicated that the mechanical performance of concrete enhanced with HH inclusion up to 3%, after which a deterioration ensued, presumably due to inadequate dispersion and workability challenges. The ANN models precisely predicted mechanical outcomes, while the RSM-derived models demonstrated strong correlations, with R2 values of 0.9434, 0.9365, and 0.9311 for CS, FS, and STS, respectively. ANOVA confirmed the significance of model inputs with p-values below 0.05. Furthermore, SEM, EDX, and XRD analyses validated the integration of HH into the concrete matrix and substantiated the observed mechanical properties. This study confirms the feasibility of HH as a sustainable fiber in concrete, enhancing critical performance metrics when applied at optimal dosages. The amalgamation of ANN, RSM, and ANOVA offers a thorough methodology for optimizing innovative concrete composites and clarifying the mechanisms underlying performance enhancement.
Citation - WoS: 8
Citation - Scopus: 7
Predicting Compressive Strength of Color Pigment Incorporated Roller Compacted Concrete Via Machine Learning Algorithms: a Comparative Study
(Springernature, 2023) Çalış, Gökhan; Yildizel, Sadik Alper; Keskin, Ülkü Sultan
Due to its low albedo, traditional asphalt pavement contributes to the urban heat island effect. Color pigment added roller compacted high performance concrete is a novel approach to reducing the urban heat island effect through the use of paving materials. In this study, color pigment added roller compacted concrete specimens were produced and evaluate via the machine learning algorithms. Predicting compressive strength of concrete by utilization of machine learning methods is highly preferred method by scholars and professionals since ingredients' resources are intensive and time consuming. This research focused to predict the compressive strength of color pigment incorporated roller compacted concrete by applying multiple linear regression (ML), gradient boosting (GB), random forest (RF), support vector machines (SVM), artificial neural network (ANN) and bagging algorithms (BGG). A comprehensive database containing coarse aggregates, fine aggregate, water, cement and pigment amounts and density, age information as input parameters. The analysis results reveal that Bagging algorithm was able to obtain more satisfactory results than the other algorithms in predicting compressive strength (CS) of color pigment incorporated roller compacted concrete. In this algorithm, root mean square error (RMSE) was determined to be 1.53, R-2 to be 0.962, mean absolute error (MAE) to be 0.916, and mean absolute percentage error (MAPE) to be 0.033. ANN algorithm showed significant accuracy in prediction process with RMSE of 1.725, R-2 of 0.949, MAE of 1.144, and MAPE of 0.040. The lowest accuracy was obtained in SVM algorithm with RMSE of 26.910 R-2 of 0.512, MAE of 3.981, and MAPE of 0.040. Therefore, the present study can provide an efficient option for estimating the of color added Roller compacted concrete for pavements.
Retraction: Application of Waste Ceramic Powder as a Cement Replacement in Reinforced Concrete Beams Toward Sustainable Usage in Construction
(Elsevier, 2024) Aksoylu, Ceyhun; Ozkilic, Yasin Onuralp; Bahrami, Alireza; Yildizel, Sadik Alper; Hakeem, Ibrahim Y.; Ozdoner, Nebi; Karalar, Memduh
[No Abstract Available]
Sustainable Concrete with Waste Tire Rubber and Recycled Steel Fibers: Experimental Insights and Hybrid PINN-CatBoost Prediction
(MDPI, 2025) Ecemis, Ali Serdar; Yildizel, Sadik Alper; Beskopylny, Alexey N.; Stel'makh, Sergey A.; Shcherban', Evgenii M.; Aksoylu, Ceyhun; Ozkilic, Yasin Onuralp
The growing environmental concern over waste tire accumulation necessitates innovative recycling strategies in construction materials. Therefore, this study aims to develop and evaluate sustainable concrete by integrating waste tire rubber (WTR) aggregates of different sizes and recycled waste tire steel fibers (WTSFs), assessing their combined effects on the mechanical and microstructural performance of concrete through experimental and analytical approaches. WTR aggregates, consisting of fine (0-4 mm), small coarse (5-8 mm), and large coarse (11-22 mm) particles, were used at substitution rates of 0-20%; WTSF was used at volumetric dosages of 0-2%, resulting in a total of 40 mixtures. Mechanical performance was evaluated using density and pressure resistance tests, while microstructural properties were assessed using scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). The findings indicate systematic decreases in density and compressive strength with increasing WTR ratio; the average strength losses were approximately 12%, 20%, and 31% at 5%, 10%, and 20% for WTR substitution, respectively. Among the WTR types, the most negative effect occurred in fine particles (FWTR), while the least negative effect occurred in coarse particles (LCWTR). The addition of WTSF compensated for losses at low/medium dosages (0.5-1.0%) and increased strength by 2-10%. However, high dosages (2.0%) reduced strength by 20-40% due to workability issues, fiber clumping, and void formation. The highest strength was achieved in the 5LCWTR-1WTSF mixture at 36.98 MPa (approximate to 6% increase compared to the reference/control concrete), while the lowest strength was measured at 16.72 MPa in the 20FWTR-2WTSF mixture (approximate to 52% decrease compared to the reference/control). A strong positive correlation was found between density and strength (r, Pearson correlation coefficient, approximate to 0.77). SEM and EDX analyses confirmed the weak matrix-rubber interface and the crack-bridging effect of steel fibers in mixtures containing fine WTR. Additionally, a hybrid prediction model combining physics-informed neural networks (PINNs) and CatBoost, supported by data augmentation strategies, accurately estimated compressive strength. Overall, the results highlight that optimized integration of WTR and WTSF enables sustainable concrete production with acceptable mechanical and microstructural performance.