BETONARME DÖŞEMELERİN ÇARPMA DAVRANIŞI ÜZERİNE İSTATİSTİKSEL ÇALIŞMA

Öz

Betonarme döşemeler bir binanın taşıyıcı sisteminde önemli elemanlardır. Betonarme döşemeler statik ölü ve hareketli yüklerin etkisi altında tasarlanmalarına rağmen hizmet ömürleri boyunca düşük hızlı darbe yüklerine maruz kalabilirler. Bu çalışmada öncelikle basit mesnetli betonarme bir döşemenin ani darbe etkisi altındaki davranışı deneysel olarak incelenmiştir. Bu amaçla, darbe deneylerinde ağırlık düşürme testi düzeneği, ivmeölçer, dinamik yük hücresi, optik fotosel ve veri kaydedici gibi temel ölçüm cihazları kullanılmaktadır. Sayısal analiz bölümünde, betonarme döşemelerin sentetik verilerini üretmek için Çekişmeli Üretici Ağları (GANs) kullanan yeni bir yöntem önerilmiştir. GAN'lar, standart denetimli veya denetimsiz öğrenme tekniklerinin ötesine geçen bir tür derin öğrenme algoritmasıdır. Bunlar, mümkün olduğu kadar özgün, doğal bir veri kümesine en yakın verileri üretmek için iki ağın birbiriyle rekabet ettiği bir tür üretici modeldir. Bir ağ üretici olarak adlandırılır ve ham girdileri alır. Diğer ağ, üretilen çıktıları alan ve bunların sahte olup olmadığını belirlemeye çalışan ayırıcı olarak adlandırılır. Sayısal analiz bulguları, modelimizin, betonarme döşemenin gerçek davranışını düzgün bir şekilde temsil eden sentetik verilerini üretme yeteneğine sahip olduğunu göstermektedir. Bu durum, fiziksel testler yapmak zorunda kalmadan analizlere girdi için veri üretme imkânı sağlayacağından, betonarme döşemeler üzerinde dinamik analiz yapması gereken mühendisler için değerli bir araç olabilir.

Anahtar Kelimeler

• Darbe
• döşeme
• test düzeneği
• GANs

STATISTICAL STUDY ON IMPACT BEHAVIOR OF A REINFORCED CONCRETE SLAB

Öz

Reinforced concrete (RC) slabs are significant members in the structural system of a building. Although RC slabs are designed under the effect of static dead and live loads, they may be exposed to low velocity impact loading during their service lives. In this study, behavior of a simply supported RC slab under sudden impact effect is experimentally investigated in the first place. For this purpose, a drop weight test setup and essential measurement devices such as accelerometers, dynamic load cell, optic photocells, and data logger are utilized in impact experiments. In the numerical analysis part, a new method is proposed which uses the Generative Adversarial Networks (GANs) for generating the synthetic data of RC slabs. GANs are a type of deep learning algorithm that goes beyond the standard supervised or unsupervised learning techniques. They're a type of generative model in which two networks compete against each other to generate data that is as close to an authentic, natural dataset as possible. One network is called the generator and takes in raw inputs. The other network is called the discriminator, which takes in the generated outputs and tries to identify them as fake or not. The numerical analysis findings show that our model is capable of generating tabular synthetic data that properly represent the RC slab's actual behavior. This situation could be a valuable tool for engineers who need to perform dynamic analysis on RC slabs, as it would allow them to generate data for input into their analysis without having to perform physical tests.

Anahtar Kelimeler

• Impact
• slab
• test setup
• GANs

Kaynakça

1. E23-00, A. (2002). Standard test methods for notched bar impact testing of metallic materials. ASTM International.
2. Erdem, R. T. (2021). Dynamic responses of reinforced concrete slabs under sudden impact loading. Revista de La Construcción, 20(2), 346–358.
3. Erdem, R. T. (2014). Prediction of acceleration and impact force values of a reinforced concrete slab. Computers and Concrete, 14(5), 563–575.
4. Erdem, R. T. & Berberoğlu, M. (2021). Prediction of impact results on cement based mortar slabs. Romanian Journal of Materials, 34(9), 1517–1534.
5. Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. & Bengio, Y. (2014). Generative Adversarial Networks.
6. Grnarova, P., Levy, K. Y., Lucchi, A., Perraudin, N., Goodfellow, I., Hofmann, T. & Krause, A. (2019). A domain agnostic measure for monitoring and evaluating GANs. Advances in Neural Information Processing Systems, 32(NeurIPS).
7. Hsu, C.-C., Zhuang, Y.-X. & Lee, C.-Y. (2020). Deep Fake Image Detection Based on Pairwise Learning. In Applied Sciences (Vol. 10, Issue 1). https://doi.org/10.3390/app10010370
8. Hummeltenberg, A., Beckmann, B., Weber, T. & Curbach, M. (2011). Investigation of Concrete Slabs under Impact Load. Applied Mechanics and Materials, 82. https://doi.org/10.4028/www.scientific.net/AMM.82.398

9. Iqbal, M. A., Kumar, V. & Mittal, A. K. (2019). Experimental and numerical studies on the drop impact resistance of prestressed concrete plates. International Journal of Impact Engineering.
10. Khalilpourazari, S., Khalilpourazary, S., Özyüksel Çiftçioğlu, A. & Weber, G.-W. (2021). Designing energy-efficient high-precision multi-pass turning processes via robust optimization and artificial intelligence. Journal of Intelligent Manufacturing, 32(6), 1621–1647. https://doi.org/10.1007/s10845-020-01648-0
11. Kumar, V., Iqbal, M. A. & Mittal, A. K. (2018). Study of induced prestress on deformation and energy absorption characteristics of concrete slabs under drop impact loading. Construction and Building Materials, 188(Complete), 656–675. https://doi.org/10.1016/j.conbuildmat.2018.08.113
12. LeCun, Y., Bengio, Y. & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539
13. Mozo, A., González-Prieto, Á., Pastor, A., Gómez-Canaval, S. & Talavera, E. (2022). Synthetic flow-based cryptomining attack generation through Generative Adversarial Networks. Scientific Reports, 12(1), 2091. https://doi.org/10.1038/s41598-022-06057-2
14. Navidan, H., Moshiri, P. F., Nabati, M., Shahbazian, R., Ghorashi, S. A., Shah-Mansouri, V. & Windridge, D. (2021). Generative Adversarial Networks (GANs) in networking: A comprehensive survey & evaluation. Computer Networks, 194, 108149. https://doi.org/https://doi.org/10.1016/j.comnet.2021.108149
15. Othman, H. & Marzouk, H. (2015). An experimental investigation on the effect of steel reinforcement on impact response of reinforced concrete plates. International Journal of Impact Engineering, 88. https://doi.org/10.1016/j.ijimpeng.2015.08.015
16. Rather, S. A. & Bala, P. S. (2020). Swarm-based chaotic gravitational search algorithm for solving mechanical engineering design problems. World Journal of Engineering, 17(1), 97–114. https://doi.org/10.1108/WJE-09-2019-0254
17. Rather, S. A. & Bala, P. S. (2021). Lévy flight and chaos theory-based gravitational search algorithm for mechanical and structural engineering design optimization. Open Computer Science, 11(1), 509–529. https://doi.org/doi:10.1515/comp-2020-0223
18. Xiao, Y., Li, B. & Fujikake, K. (2017). Behavior of reinforced concrete slabs under low-velocity impact. ACI Structural Journal, 114(3), 643–658. https://doi.org/10.14359/51689565
19. Xu, L., Skoularidou, M., Cuesta-Infante, A. & Veeramachaneni, K. (2019). Modeling Tabular data using Conditional GAN. Advances in Neural Information Processing Systems.
20. Yılmaz, T., Anil, Ö. & Erdem, R. T. (2022). Experimental and numerical investigation of impact behavior of RC slab with different opening size and layout. Structures, 35, 818–832.
21. Yılmaz, Tolga, Kıraç, N., Anil, Ö., Erdem, R. T. & Kaçaran, G. (2020). Experimental Investigation of Impact Behaviour of RC Slab with Different Reinforcement Ratios. KSCE Journal of Civil Engineering, 24(1), 241–254. https://doi.org/10.1007/s12205-020-1168-x
22. Zineddin, M. & Krauthammer, T. (2007). Dynamic response and behavior of reinforced concrete slabs under impact loading. International Journal of Impact Engineering, 34, 1517–1534.