3D-DCGAN and 3D-CNN-U-Net for predicting shrinkage stresses and displacements in monolithic reinforced concrete slabs on a base

Abstract

This study presents an approach that demonstrates the capabilities of Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) in solving mechanics-related problems, particularly in the design of monolithic reinforced concrete slabs on a base. For the first time, a voxel-based representation of the studied object is proposed. In many cases, the design stage involves the inclusion of technological holes of various shapes, and the slab surface may have complex geometry. Determining the stress–strain state (SSS) using closed-form solutions under such conditions is highly labor-intensive or even unattainable. This paper presents an alternative approach using a 3D CNN with a U-Net architecture, Deep Convolutional Generative Adversarial Nets (3D-DCGAN), and an Improved GAN (I-GAN). This method enables accurate prediction of shrinkage stresses and displacements in slabs more efficiently than the finite element method (FEM). The paper highlights the promising potential of neural networks in structural engineering.

Keywords

Convolutional Neural Networks (CNNs) / Generative Adversarial Networks (CANs) / Neurons / Slabs on base / Voxels / Shrinkage

Highlight

1.	To illustrate the potential of artificial intelligence in mechanics.
2.	To demonstrate the application of soft computing with deep learning in design-related tasks.
3.	To show the advantages of neural networks in predicting forced displacements and stresses in reinforced concrete slabs on a base.
4.	To develop a database of slabs for training neural networks, with subsequent integration of available data and data planned to be generated in future stages.

The ultimate goal of the research is to develop a slab design method that combines the advantages of theoretical models, the finite element method, and bio-inspired technologies.

References

[1] Bacchus, B., & Barua, B. Provincial health index 2013. Report. Fraser Institute, January 2013, p. 25.

[2] Bamiyo SP, Uche OA, Adamu M. Prediction of load deflection behaviour of two way RC slab using neural network approach. International Journal of Optimization in Civil Engineering, 2017, 7(4): 633-644

[3] Barrodale I. L1 approximation and the analysis of data. Journal of the Royal Statistical Society Series C: Applied Statistics, 1968, 17(1): 51-57

[4] CEN. (2023). EN 1992-1-1:2023—Eurocode 2: Design of concrete structures—Part 1-1: General rules and rules for buildings. European Committee for Standardization.

[5] Fédération internationale du béton (Fib). Fib model code for concrete structures 2010, 2013Ernst & Sohn

[6] Garrett, J. H. (1992). Neural networks and their applicability within civil engineering. In: Computing in civil engineering and geographic information systems symposium. ASCE, Reston, VA, pp 1155-1162.

[7] Гoлoвкo, B. A., & Кpacнoпpoшин, B. B. (2017). Heйpoceтeвыe тexнoлoгии oбpaбoтки дaнныx.

[8] Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2015). Generative adversarial net. Departement d’informatique et de recherche operationnelle ‘Universite de Montreal’ Montreal, QC H3C 3J7.

[9] Ioffe, S., & Szegedy, C. (2015, June). Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International conference on machine learning (pp. 448–456).

[10] Isola, P., Zhu, J. Y., Zhou, T., & Efros, A. A. (2017). Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1125–1134).

[11] Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.

[12] Kruger SE, Leddy J, Howell EC, Madireddy S, Akcay C, Bechtel Amara T, McClenaghan J, Lao LL, Orozco D, Smith SP, Sun X, Samaddar A, Pankin A-Y. Thinking Bayesian for plasma physicists. Physics of Plasmas, 2024, 31(5050901

[13] Lagaros ND, Charmpis DC, Papadrakakis M. An adaptive neural network strategy for improving the computational performance of evolutionary structural optimization. Computer Methods in Applied Mechanics and Engineering, 2005, 19430-33): 3374-3393.

[14] Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3431–3440).

[15] Mиxaйлoв, B. B., & Литвep, C. Л. (1974). Pacшиpяющиe и нaпpягaющиe цeмeнты и caмoнaпpяжeнныe кoнcтpyкции. M.: Cтpoйиздaт. https://studfile.net/preview/19412330/

[16] Ministry of Architecture and Construction of the Republic of Belarus. (2020). SN 5.09.01-2020. Floors. Construction Standards. Minsk.

[17] Pettersson, D. (1998). Stresses in concrete structures from ground restraint. Licentiate thesis, Stockholm, 1998. Part II. 63 pp.

[18] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention—MICCAI 2015: 18th International conference, Munich, Germany, October 5–9, 2015, proceedings, part III 18 (pp. 234–241). Springer.

[19] Rozycki, D. K., & Rasmussen, R. O. (1998, June). Assessment of slab-base interaction in PCC airfield pavements. In Airport facilities: Innovations for the next century. Proceedings of the 25th international air transportation conference. American Society of Civil Engineers. http://worldcat.org/isbn/0784403511

[20] Selcuk S, Tang P. A metaheuristic-guided machine learning approach for concrete strength prediction with high mix design variability using ultrasonic pulse velocity data. Developments in the Built Environment, 2023, 15100220

[21] Stott JP. Tests on materials for use in sliding layers under concrete road slabs. Civil Engineering, 1961, 56(663): 1297

[22] Tully, S. H. (1997). A neural network approach for predicting the structural behavior of concrete slabs. Doctoral dissertation, Memorial University of Newfoundland. http://research.library.mun.ca/id/eprint/5319

[23] Wesevich, J. W., McCullough, B. F., & Burns, N. H. (1987). Stabilized subbase friction study for concrete pavements.

[24] Wimsatt, A. J., McCullough, B. F., & Burns, N. H. (1987). Methods of analyzing and factors influencing frictional effects of subbases.

[25] Жeлткoвич, A. E. (2009). O нaзнaчeнии пapaмeтpoв pacчeтнoй мoдeли coбcтвeнныx дeфopмaций плит пpи взaимoдeйcтвии c ocнoвaниeм. https://rep.bstu.by/handle/data/8908

[26] Жeлткoвич, A. E., & Typ, B. B. (2011). Pacчёт вынyждeнныx пepeмeщeний и нaпpяжeний oт ycaдки в мoнoлитныx бeтoнныx плитax, взaимoдeйcтвyющиx c ocнoвaниeм. https://rep.bstu.by/handle/data/4719

[27] Zadeh LAZadeh LA. Fuzzy logic, neural networks, and soft computing. Fuzzy sets, fuzzy logic, and fuzzy systems: Selected papers, 1996World Scientific Publishing775-782.

[28] Zhang J, Li VC. Influence of supporting base characteristics on shrinkage-induced stresses in concrete pavements. Journal of Transportation Engineering, 2001, 127(6): 455-462.

[29] Zheltkovich, A. E., Parkhots, K. G., Molosh, V. V., Jin, H., & Xu, S. (2023). 2D convolutional neural network in the design of monolithic self-stressed slabs on base. Vestnik of Brest State Technical University, no. 3(132), pp. 54–60.

[30] Zheltkovich, A. E., & Tur, V. V. Ustroystvo dlya polucheniya diagramm sdviga betona po sypuchim i skol’zjashchim osnovanijam. Patent no. 4080, Republic of Belarus, filed April 24, 2007, and issued December 30, 2007. Ofitsial‘nyy byulleten’, no. 6(59), 2007, p. 213.

Funding

International Cooperation Fund Project(NOS. ICR2305)

RIGHTS & PERMISSIONS

The Author(s)