Comparison of Machine Learning Classifiers for Reducing Fitness Evaluations of Structural Optimization

Document Type : Regular Article

Authors

Faculty of Building and Industrial Constructions, Hanoi University of Civil Engineering, Hanoi, Vietnam

Abstract

Metaheuristic algorithms have been widely used to solve structural optimization problems. Despite their powerful search capabilities, these algorithms often require a large number of fitness evaluations. Constructing a machine learning classifier to identify which individuals should be evaluated using the original fitness evaluation is a great solution to reduce the computational cost. However, there is still a lack of a thorough comparison between machine learning classifiers when integrating into the optimization process. This paper aims to evaluate the efficiencies of different classifiers in eliminating unnecessary fitness evaluations. For this purpose, the weight optimization of a double-layer grid structure comprising 200 members is used as a numerical experiment. Six machine learning classifiers selected for assessment in this study include Artificial Neural Network, Support Vector Machine, k-Nearest Neighbor, Decision Tree, Random Forest, and Adaptive Boosting. The comparison is made in terms of the optimal weight of the structure, the rejection rate as well as the computing time. Overall, it is found that the AdaBoost classifier achieves the best performance.

Keywords

Main Subjects


[1]     Maxwell JC. I.— On Reciprocal Figures, Frames, and Diagrams of Forces. Trans R Soc Edinburgh 1870;26:1–40. https://doi.org/10.1017/S0080456800026351.
[2]     Michelle AGM. The Limits of Economy of Material in Frame-structures, Philos Mag 1906.
[3]     Schmit LA. Structural design by systematic synthesis. Proc. Second Natl. Conf. Electron. Comput. ASCE, Sept., 1960, 1960.
[4]     Goldberg DE, Samtani MP. Engineering optimization via genetic algorithm. Electron. Comput., ASCE; 1986, p. 471–82.
[5]     Papadrakakis M, Lagaros ND, Thierauf G, Cai J. Advanced solution methods in structural optimization based on evolution strategies. Eng Comput 1998;15:12–34. https://doi.org/10.1108/02644409810200668.
[6]     Wang Z, Tang H, Li P. Optimum Design of Truss Structures Based on Differential Evolution Strategy. 2009 Int. Conf. Inf. Eng. Comput. Sci., IEEE; 2009, p. 1–5. https://doi.org/10.1109/ICIECS.2009.5365996.
[7]     Perez RE, Behdinan K. Particle swarm approach for structural design optimization. Comput Struct 2007;85:1579–88. https://doi.org/10.1016/j.compstruc.2006.10.013.
[8]     Kaveh A, Azar BF, Talatahari S. Ant Colony Optimization for Design of Space Trusses. Int J Sp Struct 2008;23:167–81. https://doi.org/10.1260/026635108786260956.
[9]     Papadrakakis M, Lagaros ND, Tsompanakis Y. Optimization of Large-Scale 3-D Trusses Using Evolution Strategies and Neural Networks. Int J Sp Struct 1999;14:211–23. https://doi.org/10.1260/0266351991494830.
[10]   Salajegheh E, Gholizadeh S. Optimum design of structures by an improved genetic algorithm using neural networks. Adv Eng Softw 2005;36:757–67. https://doi.org/10.1016/j.advengsoft.2005.03.022.
[11]   Kaveh A, Gholipour Y, Rahami H. Optimal Design of Transmission Towers Using Genetic Algorithm and Neural Networks. Int J Sp Struct 2008;23:1–19. https://doi.org/10.1260/026635108785342073.
[12]   Salajegheh E, Salajegheh J, SEYEDPOUR SM, Khatibinia M. Optimal design of geometrically nonlinear space trusses using an adaptive neuro-fuzzy inference system 2009.
[13]   Chen T-Y, Cheng Y-L. Data-mining assisted structural optimization using the evolutionary algorithm and neural network. Eng Optim 2010;42:205–22. https://doi.org/10.1080/03052150903110942.
[14]   Nguyen T-H, Vu A-T. Using Neural Networks as Surrogate Models in Differential Evolution Optimization of Truss Structures, 2020, p. 152–63. https://doi.org/10.1007/978-3-030-63007-2_12.
[15]   Rosso MM, Cucuzza R, Di Trapani F, Marano GC. Nonpenalty Machine Learning Constraint Handling Using PSO-SVM for Structural Optimization. Adv Civ Eng 2021;2021:1–17. https://doi.org/10.1155/2021/6617750.
[16]   Nguyen T-H, Vu A-T. Application of Artificial Intelligence for Structural Optimization, 2022, p. 1052–64. https://doi.org/10.1007/978-981-16-3239-6_82.
[17]   Kim S-E, Vu Q-V, Papazafeiropoulos G, Kong Z, Truong V-H. Comparison of machine learning algorithms for regression and classification of ultimate load-carrying capacity of steel frames. Steel Compos Struct 2020;37:193–209.
[18]   Nguyen T-H, Vu A-T. A Comparative Study of Machine Learning Algorithms in Predicting the Behavior of Truss Structures, 2021, p. 279–89. https://doi.org/10.1007/978-981-15-7527-3_27.
[19]   Nguyen T-H, Vu A-T. Evaluating structural safety of trusses using Machine Learning. Frat Ed Integrità Strutt 2021;15:308–18. https://doi.org/10.3221/IGF-ESIS.58.23.
[20]   Ho-Huu V, Nguyen-Thoi T, Vo-Duy T, Nguyen-Trang T. An adaptive elitist differential evolution for optimization of truss structures with discrete design variables. Comput Struct 2016;165:59–75. https://doi.org/10.1016/j.compstruc.2015.11.014.
[21]   Freund Y, Schapire RE. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J Comput Syst Sci 1997;55:119–39. https://doi.org/10.1006/jcss.1997.1504.
[22]   Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res 2011;12:2825–30.
[23]   Gholizadeh S. Optimal design of double layer grids considering nonlinear behaviour by sequential grey wolf algorithm. Iran Univ Sci Technol 2015;5:511–23.
  • Receive Date: 23 September 2021
  • Revise Date: 05 November 2021
  • Accept Date: 22 November 2021
  • First Publish Date: 22 November 2021