Metakaolin ikameli beyaz çimento harçlarının basınç dayanımlarının YSA ve ANFIS modelleriyle tahmini

Yıl 2025, Cilt: 6 Sayı: 2, 23 - 36, 30.12.2025

Azime Subaşı , Mehmet Emiroğlu , Yılmaz Koçak

https://doi.org/10.53608/estudambilisim.1818216

Öz

Bu çalışmada, yüksek reaktiviteye sahip metakaolinin, beyaz çimento harçlarının basınç dayanımı üzerine etkisinin tahmininde yapay sinir ağları (YSA) ve uyarlamalı ağ tabanlı bulanık çıkarım sistemi (ANFIS) ile oluşturulan farklı modeller kullanılmıştır. Bu sürecin ilk aşamasında beyaz Portland çimento (BPÇ) içerisine, ağırlıkça %0, %5, %10, %15, %20, %25 ve %30 oranında metakaolin (MK) ikame edilerek 7 farklı çimento hazırlanmıştır. İkinci aşamada bu çimentolar ile harçlar üretilmiş ve bu harçların basınç dayanımları 2, 7, 28, 56 ve 90. günlerde tespit edilmiştir. Üçüncü aşamada harçlarının basınç dayanımları, ileri beslemeli geri yayılımlı (YSA-1) ve Elman geri yayılımlı (YSA-2) sinir ağı modelleri ile MATLAB programındaki bulanık mantık araç kutusu içerisindeki üçgen (ANFIS-1) ve Gauss (ANFIS-2) üyelik fonksiyonları kullanılarak tahmin edilmiştir. Son etapta bu modellerin güvenilirliğinin belirlenmesi için belirleme katsayısı (R2), ortalama mutlak yüzde hata (MAPE), hata kareleri ortalamasının karekökü (RMSE) ve saçılma indeksi (SI) gibi metrikler kullanılmıştır. Gerçekleştirilen istatistiksel analizlere göre test aşamasında tahmin edilen basınç dayanımlarının gerçeğe çok yakın sonuçlar veren hem YSA-1 (R2=0.9997, ve MAPE=%0.0279, RMSE=0.2425, SI=0.0052) hem de YSA-2 modeliyle (R2=0.9989, ve MAPE=%0.0073, RMSE=0.2824, SI=0.0060) başarılı bir şekilde tahmin edilebileceği tespit edilmiştir. Dolayısıyla, istatistiksel metriklere göre tahmin edilen sonuçlar ile gerçek sonuçlar arasında iyi bir uyum olduğu söylenebilir.

Anahtar Kelimeler

Beyaz Portland çimento , Basınç dayanımı , YSA , ANFIS

Etik Beyan

Yazarlar çalışmayla ilgili herhangi bir kurum ya da kişilerle çıkar çatışmasının olmadığını beyan eder.

Teşekkür

Yazarlar, analiz ve deneylerin yapılmasında katkı sunan Mersin ÇİMSA Çimento Fabrikası çalışanlarına ve Yetkililerine teşekkürlerini sunarlar.

Estimation of compressive strength of metakaolin substituted white cement mortars using ANN and ANFIS models

Öz

In this study, different models created with ANN and ANFIS were used to predict the effect of high reactivity metakaolin on the compressive strength of white cement mortars. In the first stage of this process, seven different cements were prepared by substituting 0%, 5%, 10%, 15%, 20%, 25% and 30% by weight metakaolin (MK) into white Portland cement (WPC). In the second stage, mortars were produced with these cements and their compressive strengths were determined on days 2, 7, 28, 56 and 90. In the third stage, the compressive strengths of the mortars were estimated using feedforward backpropagation (ANN-1) and Elman backpropagation (ANN-2) neural network models and triangular (ANFIS-1) and Gaussian (ANFIS-2) membership functions in the fuzzy logic toolbox in MATLAB. In the final stage, metrics such as coefficient of determination (R2), mean absolute percentage error (MAPE), root mean square error (RMSE), and scattering index (SI) were used to determine the reliability of these models. According to the statistical analyses performed, it was determined that the compressive strengths predicted during the test phase could be successfully predicted by both the ANN-1 (R2=0.9997, MAPE=0.0279%, RMSE=0.2425, SI=0.0052) and the ANN-2 models (R2=0.9989, MAPE=0.0073%, RMSE=0.2824, SI=0.0060), which yielded results very close to reality. Therefore, it can be said that there is a good agreement between the predicted results according to statistical metrics and the actual results.

Anahtar Kelimeler

White Portland cement , Compressive strength , ANN , ANFIS

Etik Beyan

The authors declare that they have no conflict of interest with any institution or person related to the study.

Teşekkür

The authors would like to thank the employees and authorities of Mersin ÇİMSA Cement Factory for their contributions to the analysis and experiments.

Kaynakça

• Rose, D., Shirzad, S. 2024. Innovations in Green Concrete: Combining Metakaolin and Arundo Grass Biochar for Enhanced Sustainability. Sustainability, 16(24), 11219. https://doi.org/10.3390/su162411219
• Okoji, A. I., Anozie, A. N., Omoleye, J. A. 2022. Evaluating the thermodynamic efficiency of the cement grate clinker cooler process using artificial neural networks and ANFIS. Ain Shams Engineering Journal, 13(5), 101704. https://doi.org/10.1016/j.asej.2022.101704
• Yanguatin, H., Tobón, J., Ramírez, J. 2017. Pozzolanic reactivity of kaolin clays, a review. Revista Ingeniería de Construcción, 32(2), 13-24. https://doi.org/10.4067/S0718-50732017000200002
• Shafiq, N., Nuruddin, M. F., Khan, S. U., Ayub, T. 2015. Calcined kaolin as cement replacing material and its use in high strength concrete. Construction and Building Materials, 81, 313-323. https://doi.org/10.1016/j.conbuildmat.2015.02.050
• Elahi, T. E., Romero, P., Garg, N. 2026. Impact of calcination time on metakaolin morphology and enhancement of pozzolanic reactivity: Insights from in situ TEM. Cement and Concrete Research, 199, 108022. https://doi.org/10.1016/j.cemconres.2025.108022
• Raheem, A. A., Abdulwahab, R., Ikotun, B. D., Adetoro, E. A. 2024. Evaluation of the strength performance and microstructural properties of different based metakaolin blended cements containing greenly synthesized nanosilica. Discover Applied Sciences, 6(5), 239. https://doi.org/10.1007/s42452-024-05890-6
• Said-Mansour, M., Kadri, E. H., Kenai, S., Ghrici, M., Bennaceur, R. 2011. Influence of calcined kaolin on mortar properties. Construction and building Materials, 25(5), 2275-2282. https://doi.org/10.1016/j.conbuildmat.2010.11.017
• Zhang, M. H., Malhotra, V. M. 1995. Characteristics of a thermally activated alumino-silicate pozzolanic material and its use in concrete. Cement and concrete research, 25(8), 1713-1725. https://doi.org/10.1016/0008-8846(95)00167-0
• Sabir, B. B., Wild, S., Bai, J. 2001. Metakaolin and calcined clays as pozzolans for concrete: a review. Cement and concrete composites, 23(6), 441-454. https://doi.org/10.1016/S0958-9465(00)00092-5
• Siddique, R., Klaus, J. 2009. Influence of metakaolin on the properties of mortar and concrete: A review. Applied Clay Science, 43(3-4), 392-400. https://doi.org/10.1016/j.clay.2008.11.007
• Khatib, J. M., Wild, S. 1998. Sulphate resistance of metakaolin mortar. Cement and Concrete research, 28(1), 83-92. https://doi.org/10.1016/S0008-8846(97)00210-X
• Khatib, J. M., Clay, R. M. 2004. Absorption characteristics of metakaolin concrete. Cement and Concrete Research, 34(1), 19-29. https://doi.org/10.1016/S0008-8846(03)00188-1
• Kim, H. S., Lee, S. H., Moon, H. Y. 2007. Strength properties and durability aspects of high strength concrete using Korean metakaolin. Construction and Building Materials, 21(6), 1229-1237. https://doi.org/10.1016/j.conbuildmat.2006.05.007
• Loureiro, A. A. B., Stefani, R. 2024. Comparing the performance of machine learning models for predicting the compressive strength of concrete. Discover Civil Engineering, 1(1), 19. https://doi.org/10.1007/s44290-024-00022-w
• Mosquera, C. H., Acosta, M. P., Rodríguez, W. A., Mejía‐España, D. A., Torres, J. R., Martinez, D. M., Abellán‐García, J. 2025. ANN‐based analysis of the effect of SCM on recycled aggregate concrete. Structural Concrete, 26(2), 1435-1454. https://doi.org/10.1002/suco.202400024
• Topçu, İ. B., Gülbandılar, E., Koca, A. B. 2018. Reaktif Pudra Betonlarının Basınç Dayanımının ANFIS ile Tahmini. Politeknik Dergisi, 21(1), 165-171. https://doi.org/10.2339/politeknik.385923
• Erdem, Y., Gülbandılar, E., Koçak, Y. 2025. Silis dumanı ikameli çimentoların YSA ve ANFIS modelleri ile basınç dayanımlarının tahmini. Uludağ Üniversitesi Mühendislik Fakültesi Dergisi, 30(2), 355-374. https://doi.org/10.17482/uumfd.1639035
• Verma, N. K., Meesala, C. R., Kumar, S. 2023. Developing an ANN prediction model for compressive strength of fly ash-based geopolymer concrete with experimental investigation. Neural Computing and Applications, 35(14), 10329-10345. https://doi.org/10.1007/s00521-023-08237-1
• Sabour, M. R., Dezvareh, G. A., Niavol, K. P. 2021. Application of artificial intelligence methods in modeling corrosion of cement and sulfur concrete in sewer systems. Environmental Processes, 8, 1601-1618. https://doi.org/10.1007/s40710-021-00542-y
• Ramachandra, R., Mandal, S. 2023. Prediction of fly ash concrete type using ANN and SVM models. Innovative Infrastructure Solutions, 8(1), 47. https://doi.org/10.1007/s41062-022-01014-4
• Sharma, N., Thakur, M. S., Kumar, R., Malik, M. A., Alahmadi, A. A., Alwetaishi, M., Alzaed, A. N. 2022. Assessing waste marble powder impact on concrete flexural strength using Gaussian process, SVM, and ANFIS. Processes, 10(12), 2745. https://doi.org/10.3390/pr10122745
• Kocak, B., Pınarcı, İ., Güvenç, U., Kocak, Y. 2023. Prediction of compressive strengths of pumice-and diatomite-containing cement mortars with artificial intelligence-based applications. Construction and Building Materials, 385, 131516. https://doi.org/10.1016/j.conbuildmat.2023.131516
• Ozcan, G., Kocak, Y., Gulbandilar, E. 2018. Compressive strength estimation of concrete containing zeolite and diatomite: an expert system implementation. Computers and Concrete, 21(1), 21-30. https://doi.org/10.12989/cac.2018.21.1.021
• TS EN 197-1 2012. Çimento- Bölüm 1: Genel Çimentolar Bileşim, Özellikler ve Uygunluk Kriterleri. Türk Standartları, Ankara, Türkiye.
• TS EN 196-1 2016. Çimento deney metotları - Bölüm 1: Dayanım tayini. Türk Standartları, Ankara, Türkiye.
• Deore C., Aswaı S.J. 2024. A Comprehensive Review ofthe Artificial Neural Networks (ANN) Methodology Implementations for Analysing and Forecasting the Efficiency of Solar Water Heater Collector Under Various Tilt Angles. International Journal of Intelligent Systems and Applications in Engineering, 12(21), 1516-1533.
• Adesanya, E., Aladejare, A., Adediran, A., Lawal, A., Illikainen, M. 2021. Predicting shrinkage of alkali-activated blast furnace-fly ash mortars using artificial neural network (ANN). Cement and Concrete Composites, 124, 104265. https://doi.org/10.1016/j.cemconcomp.2021.104265
• Kars, F., Ozcan, G., Gulbandilar, E., Kocak, Y. 2024. Prediction of Compressive Strength and Design Parameters of C30/37, C35/45 and C40/50 Concrete Classes by ANN. Journal of Civil Engineering and Urbanism, 14(4), 356-367. https://dx.doi.org/10.54203/jceu.2024.00
• Chen, Q., Ma, N. 2024. Heart disease prediction method based on ANN. Highlights in Science, Engineering and Technology, 85, 411-417. https://doi.org/10.54097/fgt46k23
• Vatankhah, R. and Ghanatian, M. 2020. Artificial neural networks and adaptive neuro-fuzzy inference systems for parameter identification of dynamic systems. Journal of Intelligent Fuzzy Systems, 39(5), 6145-6155. https://doi.org/10.3233/jifs-189085
• Yang, Y., Liu, G., Zhang, H., Zhang, Y., Yang, X. 2024. Predicting the compressive strength of environmentally friendly concrete using multiple machine learning algorithms. Buildings, 14(1), 190. https://doi.org/10.3390/buildings14010190
• Buscema, M 2002. A brief overview and introduction to artificial neural networks. Substance Use Misuse, 37(8-10), 1093-1148. https://doi.org/10.1081/JA-120004171
• Li, J., Yan, G., Abbud, L. H., Alkhalifah, T., Alturise, F., Khadimallah, M. A., Marzouki, R. 2023. Predicting the shear strength of concrete beam through ANFIS-GA–PSO hybrid modeling. Advances in Engineering Software, 181, 103475. https://doi.org/10.1016/j.advengsoft.2023.103475
• Walia, N., Singh, H., Sharma, A. 2015. ANFIS: Adaptive neuro-fuzzy inference system-a survey. International Journal of Computer Applications, 123(13), 32-38.
• Piro, N. S., Mohammed, A., Hamad, S. M., Kurda, R. 2023. RETRACTED ARTICLE: Artificial neural networks (ANN), MARS, and adaptive network-based fuzzy inference system (ANFIS) to predict the stress at the failure of concrete with waste steel slag coarse aggregate replacement. Neural Computing and Applications, 35(18), 13293-13319. https://doi.org/10.1007/s00521-023-08439-7
• Kumar, A., Arora, H. C., Kumar, K., Garg, H. 2023. Performance prognosis of FRCM-to-concrete bond strength using ANFIS-based fuzzy algorithm. Expert Systems with Applications, 216, 119497. https://doi.org/10.1016/j.eswa.2022.119497
• Kaplan, G., Öz, A., Bayrak, B., Aydın, A. C. 2023. The effect of geopolymer slurries with clinker aggregates and marble waste powder on embodied energy and high-temperature resistance in prepacked concrete: ANFIS-based prediction model. Journal of Building Engineering, 67, 105987. https://doi.org/10.1016/j.jobe.2023.105987
• Kars, F., Özcan, G., Gülbandılar, E., Koçak, Y. 2024. C16/20, C20/25 ve C25/30 Beton Sınıflarının Basınç Dayanımlarının ANFIS ve YSA ile Tahmini. Eskişehir Türk Dünyası Uygulama ve Araştırma Merkezi Bilişim Dergisi, 5(2), 37-47. https://doi.org/10.53608/estudambilisim.1561094
• Naveed, M., Hameed, A., Qureshi, M. U., Rasool, A. M. 2023. Optimization of constituent proportions for compressive strength of sustainable geopolymer concrete: a statistical approach. Results in Engineering, 20, 101575. https://doi.org/10.1016/j.rineng.2023.101575
• Koçak, B., 2022. Yapay Zeka Tabanlı Uygulamalarla Diatomit ve Pomza İkameli Çimento Harçlarının Basınç Dayanımlarının Tahmini. Düzce Üniversitesi, Lisansüstü Eğitim Enstitüsü, Yüksek Lisans Tezi, Düzce.