Research Articles | Challenge Journal of Structural Mechanics

Prediction of optimum design of welded beam design via machine learning

Yaren Aydın, Farnaz Ahadian, Gebrail Bekdaş, Sinan Melih Nigdeli


DOI: https://doi.org/10.20528/cjsmec.2024.03.001
View Counter: Abstract | 86 times | ‒ Full Article | 34 times |

Full Text:

PDF

Abstract


Design optimization is an important engineering design topic. One of the important issues in structural design is to minimize the cost. This study based on an engineering problem of Welded Beam Design aims to minimize the cost of the beam with machine learning (ML) models depending on the constraints on applied load, shear stress, bending stress and end deflection. The data set to be used in this context was created using a metaheuristic optimization algorithm. This hybrid algorithm is based on the classical Jaya algorithm by adding the student phase of Teaching Learning Based Optimization. The dataset obtained as a result of the optimization is a dataset with 1189 rows. Six different algorithms were used for prediction analyses. These are Linear Regression, Decision Tree, Elastic Net, K-Nearest Neighbour, Random Forest, and XGBoost algorithm. In the data set, load, length, and displacement are input; the design variables such as b, h, l, t and minimum cost are output. Since there is more than one output in the dataset, Multioutput Regression is applied. The performance of regression models was assessed using the Coefficient of Determination (R²), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE). According to the results obtained, the Decision Tree Model showed the best performance among the other models (R2=1, MAE=6.13e-11, RMSE=9.47e-10).


Keywords


welded beam; multioutput regression; optimization; machine learning; minimum cost

References


Akyol K, Feneaker SOF (2022). Chaotic multi-swarm particle swarm optimization for welded beam design engineering problem. Politeknik Journal, 25(4), 1645-1660.

Arslan H, Fatih Ü, Demirci M, Taşar B, Yılmaz A (2020). Estimation of Keban Dam Lake level change using ANFIS and support vector machines. Osmaniye Korkut Ata University Journal of Institute of Science and Technology, 3(2), 71-77.

Aydın Y, Bekdaş G, Nigdeli SM, Isıkdağ Ü, Kim S, Geem ZW (2023a). Machine learning models for ecofriendly optimum design of reinforced concrete columns. Applied Sciences, 13(7), 4117.

Aydın Y, Işıkdağ Ü, Bekdaş G, Nigdeli SM, Geem ZW (2023b). Use of machine learning techniques in soil classification. Sustainability, 15(3), 1-18.

Aydın Y, Cakiroglu C, Bekdaş G, Işıkdağ Ü, Kim S, Hong J, Geem ZW (2023c). Neural network predictive models for alkali-activated concrete carbon emission using metaheuristic optimization algorithms. Sustainability, 16(1), 1-20.

Aydın Y, Bekdaş G, Nigdeli SM (2023d). Reinforced concrete beam optimization via flower pollination algorithm by changing switch probability parameter. In: Vasan, P, et al. editors. Intelligent Computing and Optimization. ICO 2023. Lecture Notes in Networks and Systems, vol 852. Springer, Cham. 66-74.

Aydin Y, Nigdeli SM, Bekdaş G (2023e). Archimedes Optimization Algorithm on a Structural Optimization Problem. In: Vasant P, et al. editors. Intelligent Computing and Optimization (ICO 2023). Lecture Notes in Networks and Systems, vol 854. Springer, Cham. 173-180.

Aydin Y, Nigdeli SM, Bekdaş G (2024a). Archimedes optimization algorithm for optimization of tuned mass dampers in structural control. SSRN (https://ssrn.com/abstract=4750319), 1-8.

Aydin Y, Nigdeli SM, Bekdaş G (2024b). Parametric investigation of archimedes optimization algorithm for a structural problem. SSRN (https://ssrn.com/abstract=4750288), 1-8.

Aytekin HT (2021). The potential importance of machine learning in the context of researchers data analysis. Ufuk University Social Sciences Institute Journal, 10(19), 85-106.

Bekdaş G, Nigdeli SM (2011). Estimating optimum parameters of tuned mass dampers using harmony search. Engineering Structures, 33(9), 2716-2723.

Bekdaş G, Nigdeli SM (2013). Mass ratio factor for optimum tuned mass damper strategies. International Journal of Mechanical Sciences, 71, 68-84.

Bekdaş G (2015). Harmony search algorithm approach for optimum design of post-tensioned axially symmetric cylindrical reinforced concrete walls. Journal of Optimization Theory and Applications, 164, 342–358.

Bekdaş G, Nigdeli SM, Yang XS (2015). Sizing optimization of truss structures using flower pollination algorithm. Applied Soft Computing, 37, 322-331.

Bekdaş G, Nigdeli SM (2016). Bat algorithm for optimization of reinforced concrete columns. PAMM, 16(1), 681-682.

Bekdaş G, Cakiroglu C, Islam K, Kim S, Geem ZW (2022). Optimum design of cylindrical walls using ensemble learning methods. Applied Sciences, 12(4), 2165. 1-16.

Bekdaş G, Cakiroglu C, Kim S, Geem ZW (2022). Optimization and predictive modeling of reinforced concrete circular columns. Materials, 15(19), 6624, 1-18.

Bekdaş G, Aydın Y, Isıkdağ Ü, Sadeghifam AN, Kim S, Geem ZW (2023). Prediction of cooling load of tropical buildings with machine learning. Sustainability, 15(11), 9061. 1-20.

Bekdaş G, Cakiroglu C, Kim S, Geem ZW (2023). Optimal dimensions of post-tensioned concrete cylindrical walls using harmony search and ensemble learning with SHAP. Sustainability, 15(10), 7890, 1-18.

Cakiroglu C, Islam K, Bekdaş G, Billah M (2021). CO2 emission and cost optimization of concrete-filled steel tubular (CFST) columns using metaheuristic algorithms. Sustainability, 13(14), 8092, 1-16.

Cakiroglu C, Islam K, Bekdaş G, Isikdag U, Mangalathu S (2022). Explainable machine learning models for predicting the axial compression capacity of concrete filled steel tubular columns. Construction and Building Materials, 356, 129227, 1-20.

Cakiroglu C, Bekdaş G (2023). Predictive modeling of recycled aggregate concrete beam shear strength using explainable ensemble learning methods. Sustainability, 15(6), 4957, 1-16.

Cakiroglu C, Islam K, Bekdaş G (2023a). Manta ray foraging and jaya hybrid optimization of concrete filled steel tubular stub columns based on CO2 emission. In: Bekdaş G, Nigdeli SM (eds) Hybrid Metaheuristics in Structural Engineering. Studies in Systems, Decision and Control, vol 480. Springer, Cham. 111-125.

Cakiroglu C, Islam K, Bekdaş G, Nehdi ML (2023b). Data-driven ensemble learning approach for optimal design of cantilever soldier pile retaining walls. Structures, 51, 1268-1280.

Cao Y, Wakil K, Alyousef R, Jermsittiparsert K, Ho LS, Alabduljabbar H, Mohamed AM (2020). Application of extreme learning machine in behavior of beam to column connections. Structures, 25, 861-867.

Çoşut M, Bekdaş G, Nigdeli SM (2024). Area optimization for a T-shaped element by using pure bending equation and particle swarm optimization. SSRN (https://ssrn.com/abstract=475182), 1-8.

Çene E (2022). Predicting Euroleague basketball match outcomes with machine learning techniques and revealing the most important game related variables. Journal of Sports and Performance Research, 13(1), 31-54.

Dogan G, Arslan MH, Baykan OK (2020). Determination of damage levels of RC columns with a smart system oriented method. Bulletin of Earthquake Engineering, 18(7), 3223-3245.

Fister Jr I, Yang XS, Fister I, Brest J, Fister D (2013). A brief review of nature-inspired algorithms for optimization. Elektrotehniški vestnik, Journal of Electrical Engineering and Computer Science, 80(3), 1-7.

Huang Q, Cox RF, Shaurette M, Wang J (2012). Intelligent building hazard detection using wireless sensor network and machine learning techniques. In: Raymond Issa R, Flood I, editors. Computing in Civil Engineering, ASCE Reston VA, USA. 485-492.

Jeon JS, Shafieezadeh A, DesRoches R (2014). Statistical models for shear strength of RC beam-column joints using machine-learning techniques. Earthquake Engineering & Structural Dynamics, 43(14), 2075-2095.

Kayabekir AE, Sayın B, Nigdeli SM, Bekdaş G (2018). Jaya algorithm based optimum carbon fiber reinforced polymer design for reinforced concrete beams. AIP Conference Proceedings, 1978(1), 1-4.

Liu T, Cakiroglu C, Islam K, Wang Z, Nehdi ML (2024). Explainable machine learning model for predicting punching shear strength of FRC flat slabs. Engineering Structures, 301, 117276, 1-20.

Nguyen NH, Abellán-García J, Lee S, Nguyen TK, Vo TP (2023). Simultaneous prediction the strain and energy absorption capacity of ultra-high performance fiber reinforced concretes by using multi-output regression model. Construction and Building Materials, 384, 131418. 1-20.

Okazaki Y, Okazaki S, Asamoto S, Chun PJ (2020). Applicability of machine learning to a crack model in concrete bridges. Computer-Aided Civil and Infrastructure Engineering, 35(8), 775-792.

Osman IH, Laporte G (1996). Metaheuristics: A bibliography. Annals of Operations Research, 63, 513-623.

Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Duchesnay É (2011). Scikit-learn: Machine learning in Python. The Journal of Machine Learning Research, 12, 2825-2830.

Rao R (2016). Jaya: A simple and new optimization algorithm for solving constrained and unconstrained optimization problems. International Journal of Industrial Engineering Computations, 7(1), 19-34.

Savsani V (2014). Implementation of modified artificial bee colony (ABC) optimization technique for minimum cost design of welded structures. International Journal for Simulation and Multidisciplinary Design Optimization, 5, 1-16.

Steel Force. Welded Beams. https://www.steelforce.com.au [accessed 20-03-2024].

Ulusoy S, Niğdeli SM, Bekdaş G (2019). Optimization of PID controller parameters for active control of single degree of freedom structures. Challenge Journal of Structural Mechanics, 5(4), 130-140.

Ulusoy S, Kayabekir AE, Bekdaş G, Nigdeli SM (2020). Metaheuristic algorithms in optimum design of reinforced concrete beam by investigating strength of concrete. Challenge Journal of Concrete Research Letters, 11(2)- 26-30.

Yang XS (2010). Nature-Inspired Metaheuristic Algorithms. Luniver Press, UK.

Yücel M, Nigdeli SM, Kayabekir AE, Bekdaş G (2021). Optimization and artificial neural network models for reinforced concrete members. In: Carbas S, Toktas A, Ustun D, editors. Nature-Inspired Metaheuristic Algorithms for Engineering Optimization Applications. Springer Tracts in Nature-Inspired Computing. Springer, Singapore. 181-199.

Yücel M, Bekdaş G, Nigdeli SM, Kayabekir AE (2021). An artificial intelligence-based prediction model for optimum design variables of reinforced concrete retaining walls. International Journal of Geomechanics, 21(12), 04021244, 1-20.


Refbacks

  • There are currently no refbacks.