Abstract
Optimization of the dynamic performance of composite laminates has become necessary in many design applications of structural engineering. Although finite element analysis-based metaheuristic search is a widely adopted approach for design optimization of composite laminates, recently there has been a tremendous impetus on the use of metamodel-based metaheuristic optimization, mainly due to significant saving in computational time as well as cost. In this paper, genetic programming-based symbolic regression (SR) metamodeling technique is introduced for predicting natural frequencies of composite laminates. The optimization performance based on SR metamodels is compared with the traditionally adopted polynomial regression (PR) metamodels. The SR and PR metamodels are further solved using three metaheuristic search algorithms, i.e. genetic algorithm (GA), repulsive particle swarm optimization with local search (RPSOLC) and co-evolutionary host-parasite (CHP) for single objective optimization problems. A comprehensive analysis reveals that the SR metamodels are more accurate and compact than the PR metamodels in addition to better interpretability. Furthermore, CHP algorithm is noticed to consistently outperform both the GA and RPSOLC algorithms.
Similar content being viewed by others
Availability of data
Data available within the article or its supplementary materials.
References
Safaei, B., Chukwueloka Onyibo, E., Hurdoganoglu, D.: Effect of static and harmonic loading on the honeycomb sandwich beam by using finite element method. Facta Univ. Ser. Mech. Eng. 20(2), 279–306 (2022)
Safaei, B., Chukwueloka Onyibo, E., Goren, M., Kotrasova, K., Yang, Z., Arman, S., Asmael, M.: Free vibration investigation on RVE of proposed honeycomb sandwich beam and material selection optimization. Facta Univ. Ser. Mech. Eng. (2022). https://doi.org/10.22190/FUME220806042S
Kalita, K., Dey, P., Haldar, S.: Search for accurate RSM metamodels for structural engineering. J. Reinf. Plast. Compos. 38(21–22), 995–1013 (2019)
Hajihassani, M., Abdullah, S.S., Asteris, P.G., Armaghani, D.J.: A gene expression programming model for predicting tunnel convergence. Appl. Sci. 9(21), 4650 (2019)
Asteris, P.G., Kolovos, K.G.: Self-compacting concrete strength prediction using surrogate models. Neural Comput. Appl. 31(1), 409–424 (2019)
García-Macías, E., Castro-Triguero, R., Friswell, M.I., Adhikari, S., Sáez, A.: Metamodel-based approach for stochastic free vibration analysis of functionally graded carbon nanotube reinforced plates. Compos. Struct. 152, 183–198 (2016)
Ju, S., Shenoi, R.A., Jiang, D., Sobey, A.J.: Multi-parameter optimization of lightweight composite triangular truss structure based on response surface methodology. Compos. Struct. 97, 107–116 (2013)
Ganguli, R.: Optimum design of a helicopter rotor for low vibration using aeroelastic analysis and response surface methods. J. Sound Vib. 258(2), 327–344 (2002)
Heinonen, O., Pajunen, S.: Optimal design of stiffened plate using metamodeling techniques. J. Struct. Mech. 44(3), 218–230 (2011)
Dey, S., Mukhopadhyay, T., Khodaparast, H.H., Adhikari, S.: A response surface modelling approach for resonance driven reliability-based optimization of composite shells. Period. Polytech. Civ. Eng. 60(1), 103–111 (2016)
Todoroki, A., Suenaga, K., Shimamura, Y.: Stacking sequence optimizations using modified global response surface in lamination parameters. Adv. Compos. Mater 12(1), 35–55 (2003)
Dey, S., Mukhopadhyay, T., Adhikari, S.: Metamodel based high-fidelity stochastic analysis of composite laminates: a concise review with critical comparative assessment. Compos. Struct. 171, 227–250 (2017)
Nik, M.A., Fayazbakhsh, K., Pasini, D., Lessard, L.: A comparative study of metamodeling methods for the design optimization of variable stiffness composites. Compos. Struct. 107, 494–501 (2014)
Cevik, A., Göğüş, M.T., Güzelbey, İH., Filiz, H.: Soft computing based formulation for strength enhancement of CFRP confined concrete cylinders. Adv. Eng. Softw. 41(4), 527–536 (2010)
Singh, A.P., Mani, V., Ganguli, R.: Genetic programming metamodel for rotating beams. Comput. Model. Eng. Sci. 21(2), 133–148 (2007)
Kalita, K., Mukhopadhyay, T., Dey, P., Haldar, S.: Genetic programming-assisted multi-scale optimization for multi-objective dynamic performance of laminated composites: the advantage of more elementary-level analyses. Neural Comput. Appl. 32, 7969–7993 (2020)
Kalita, K., Dey, P., Haldar, S., Gao, X.-Z.: Optimizing frequencies of skew composite laminates with metaheuristic algorithms. Eng. Comput. 36, 741–761 (2020)
Kalita, K., Haldar, S.: Eigen frequencies of simply supported taper plates with cut-outs. Struct. Eng. Mech. 63(1), 103–113 (2017)
Kalita, K., Haldar, S.: Free vibration analysis of rectangular plates with central cutout. Cogent Eng. 3(1), 1163781 (2016)
Kalita, K., Ramachandran, M., Raichurkar, P., Mokal, S.D., Haldar, S.: Free vibration analysis of laminated composites by a nine node isoparametric plate bending element. Adv. Compos. Lett. 25(5), 108–116 (2016)
Mukhopadhyay, T., Dey, T.K., Chowdhury, R., Chakrabarti, A.: Structural damage identification using response surface-based multi-objective optimization: a comparative study. Arab. J. Sci. Eng. 40(4), 1027–1044 (2015)
Box, G.E., Cox, D.R.: An analysis of transformations. J. R. Stat. Soc. Ser. B (Methodol.) 26, 211–252 (1964)
Goldberg, D.E.: Genetic Algorithms. Pearson Education India, New Delhi (2006)
Eberhart, R., Kennedy, J.: A new optimizer using particle swarm theory. In: Proceedings of the Sixth International Symposium on Micro Machine and Human Science, 39–43 (1995)
Yang, X. S., Deb, S.: Cuckoo search via Levy flights. In: World Congress on Nature & Biologically Inspired Computing, pp. 210–214 (2009)
Shooshtari, A., Razavi, S.: A closed form solution for linear and nonlinear free vibrations of composite and fiber metal laminated rectangular plates. Compos. Struct. 92(11), 2663–2675 (2010)
Acknowledgements
The authors acknowledge the help of Dr. Ranjan Ghadai and Dr. Salil Haldar for their constitutive comments on a previous draft.
Funding
This research has received no external funding.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
No potential conflict of interest was reported by the author(s).
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A: Polynomial regression metamodels for the test problems (TP)
TP-01 | \({\left({\lambda }_{1}\right)}^{2}=219.8785+213.5809\left(\frac{{E}_{1}}{{E}_{2}}\right)+439.0622\left(\frac{{G}_{12}}{{E}_{2}}\right)+279.9152\left(\frac{{G}_{23}}{{E}_{2}}\right)+250.7809\left({ \upsilon }_{12}\right)+4.9276\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{12}}{{E}_{2}}\right)+4.5119\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{23}}{{E}_{2}}\right)-246.4965\left(\frac{{G}_{12}}{{E}_{2}}\right)\left(\frac{{G}_{23}}{{E}_{2}}\right)-252.7662\left(\frac{{G}_{23}}{{E}_{2}}\right)\left({ \upsilon }_{12}\right)-0.0551{\left(\frac{{E}_{1}}{{E}_{2}}\right)}^{2}-152.6717{\left(\frac{{G}_{12}}{{E}_{2}}\right)}^{2}-147.8168{\left(\frac{{G}_{23}}{{E}_{2}}\right)}^{2}\) |
\({\left({\lambda }_{21}\right)}^{2}=-27.926+263.398\left(\frac{{E}_{1}}{{E}_{2}}\right)+1102.0988\left(\frac{{G}_{12}}{{E}_{2}}\right)+855.4249\left(\frac{{G}_{23}}{{E}_{2}}\right)+637.1448\left({ \upsilon }_{12}\right)+13.1769\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{12}}{{E}_{2}}\right)+13.7958\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{23}}{{E}_{2}}\right)-743.6287\left(\frac{{G}_{12}}{{E}_{2}}\right)\left(\frac{{G}_{23}}{{E}_{2}}\right)-738.7692\left(\frac{{G}_{23}}{{E}_{2}}\right)\left({ \upsilon }_{12}\right)-0.1493{\left(\frac{{E}_{1}}{{E}_{2}}\right)}^{2}-423.1414{\left(\frac{{G}_{12}}{{E}_{2}}\right)}^{2}-468.0938{\left(\frac{{G}_{23}}{{E}_{2}}\right)}^{2}\) | |
TP-02 | \(ln\left({\lambda }_{1}\right)= 4.4282-0.3614\left(\frac{b}{a}\right)-1.3347\left(\frac{h}{a}\right)-1.2572\left(\alpha \right)+0.1514\left(n\right)+8.6191\left(\frac{b}{a}\right)\left(\alpha \right)-0.0742\left(\frac{h}{a}\right)\left(\alpha \right)-0.0391\left(\frac{h}{a}\right)\left(n\right)-0.0002\left(\alpha \right)\left(n\right)+0.0428{\left(\frac{b}{a}\right)}^{2}+0.0004{\left(\alpha \right)}^{2}-0.0071{\left(n\right)}^{2}\) |
\(ln\left({\lambda }_{21}\right)= 5.106-1.5926\left(\frac{b}{a}\right)-4.9656\left(\frac{h}{a}\right)-0.0055\left(\alpha \right)+5.4627\left(n\right)+1.2641\left(\frac{b}{a}\right)\left(\frac{h}{a}\right)-0.0047\left(\frac{b}{a}\right)\left(\alpha \right)-0.8106\left(\frac{h}{a}\right)\left(n\right)+0.1188{\left(\frac{b}{a}\right)}^{2}-0.0232{\left(n\right)}^{2}\) | |
TP-03 | \({\lambda }_{1}=64.7921-0.0009\left({\theta }_{1}\right)-0.0131\left({\theta }_{2}\right)-0.0011\left({\theta }_{3}\right)+0.0097\left({\theta }_{4}\right)+0.0001\left({\theta }_{2}\right)\left({\theta }_{3}\right)-0.0001\left({\theta }_{3}\right)\left({\theta }_{4}\right)-0.0029{\left({\theta }_{1}\right)}^{2}-0.0017{\left({\theta }_{2}\right)}^{2}-0.0009{\left({\theta }_{3}\right)}^{2}\) |
\({\lambda }_{21}=12.3146-0.0031\left({\theta }_{1}\right)-0.0136\left({\theta }_{2}\right)-0.0188\left({\theta }_{3}\right)-0.0263\left({\theta }_{4}\right)+0.0004\left({\theta }_{1}\right)\left({\theta }_{3}\right)+0.0003\left({\theta }_{1}\right)\left({\theta }_{4}\right)-0.0002\left({\theta }_{2}\right)\left({\theta }_{3}\right)-0.0002\left({\theta }_{2}\right)\left({\theta }_{4}\right)+0.0017{\left({\theta }_{1}\right)}^{2}+0.0008{\left({\theta }_{2}\right)}^{2}-0.0004{\left({\theta }_{3}\right)}^{2}\) | |
TP-04 | \(\sqrt{{\lambda }_{1}}=13.4425+0.0218\left(\frac{{E}_{1}}{{E}_{2}}\right)-0.0123\left(\frac{{G}_{12}}{{E}_{2}}\right)-21.0942\left(\frac{{G}_{23}}{{E}_{2}}\right)+1.3446\left({ \upsilon }_{12}\right)-1.7704\left(\frac{b}{a}\right)+22.3928\left(\frac{h}{a}\right)+0.0295\left(\alpha \right)+0.0079\left({\theta }_{1}\right)+0.0033\left({\theta }_{2}\right)- 0.0144\left({\theta }_{3}\right)+0.0289\left({\theta }_{4}\right)+0.0548\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{12}}{{E}_{2}}\right)-0.5112\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{h}{a}\right)+0.0004\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\alpha \right)-0.7894\left(\frac{{G}_{12}}{{E}_{2}}\right)\left(\frac{b}{a}\right)+0.0326\left(\frac{{G}_{12}}{{E}_{2}}\right)\left(\alpha \right)- 0.0148\left(\frac{{G}_{12}}{{E}_{2}}\right)\left({\theta }_{1}\right)+0.0162\left(\frac{{G}_{12}}{{E}_{2}}\right)\left({\theta }_{3}\right)-0.0145\left(\frac{{G}_{12}}{{E}_{2}}\right)\left({\theta }_{4}\right)+0.6889\left(\frac{{G}_{23}}{{E}_{2}}\right)\left(\frac{b}{a}\right)-0.0592\left({ \upsilon }_{12}\right)\left({\theta }_{4}\right)+1.2734\left(\frac{b}{a}\right)\left({\theta }_{3}\right)-1.7528\left(\frac{b}{a}\right)\left({\theta }_{4}\right)-1.3099\left(\frac{h}{a}\right)\left(\alpha \right)- 0.0769\left(\frac{h}{a}\right)\left({\theta }_{2}\right)+2.9684\times {10}^{-5}\left({\theta }_{3}\right)\left({\theta }_{4}\right)+18.8401\left(\frac{{G}_{23}}{{E}_{2}^{2}}\right)+0.1672{\left(\frac{b}{a}\right)}^{2}+0.0016{\left(\alpha \right)}^{2}-0.0001{\left({\theta }_{1}\right)}^{2}-0.0001{\left({\theta }_{3}\right)}^{2}\) |
\(ln\left({\lambda }_{21}\right)=1.2618+0.0275\left(\frac{{E}_{1}}{{E}_{2}}\right)+1.3682\left(\frac{{G}_{12}}{{E}_{2}}\right)+3.9339\left(\frac{{G}_{23}}{{E}_{2}}\right)+10.8273\left({ \upsilon }_{12}\right)-1.2377\left(\frac{b}{a}\right)-24.1563\left(\frac{h}{a}\right)+0.0191\left(\alpha \right)+0.0041\left({\theta }_{1}\right)+0.0047\left({\theta }_{2}\right)+0.0124\left({\theta }_{3}\right)-0.0247-0.0014\left({\theta }_{4}\right)-0.0449\left(\frac{{E}_{1}}{{E}_{2}}\right)\left(\frac{{G}_{23}}{{E}_{2}}\right)-6.7378\times {10}^{-5}\left(\frac{{E}_{1}}{{E}_{2}}\right)\left({\theta }_{2}\right)-9.7069\times {10}^{-5}\left(\frac{{E}_{1}}{{E}_{2}}\right)\left({\theta }_{3}\right)-0.0288\left(\frac{{G}_{12}}{{E}_{2}}\right)\left(\alpha \right)-13.9501\left(\frac{{G}_{23}}{{E}_{2}}\right)\left({ \upsilon }_{12}\right)+17.2227\left(\frac{{G}_{23}}{{E}_{2}}\right)\left(\frac{h}{a}\right)+0.0095\left(\frac{{G}_{23}}{{E}_{2}}\right)\left({\theta }_{4}\right)-0.0247\left({ \upsilon }_{12}\right)\left({\theta }_{3}\right)+5.1894\left(\frac{b}{a}\right)\left(\frac{h}{a}\right)-0.0014\left(\frac{b}{a}\right)\left({\theta }_{4}\right)+0.1978\left(\frac{h}{a}\right)\left(\alpha \right)-8.2466\times {10}^{-5}\left(\alpha \right)\left({\theta }_{1}\right)-5.6789\left(\alpha \right)\left({\theta }_{3}\right)+2.3161\left({\theta }_{1}\right)\left({\theta }_{4}\right)-+0.0798{\left(\frac{b}{a}\right)}^{2}-4.8291\times {10}^{-5}{\left({\theta }_{1}\right)}^{2}\) |
Appendix B: Genetic programming metamodels for the test problems
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kalita, K., Chakraborty, S. An efficient approach for metaheuristic-based optimization of composite laminates using genetic programming. Int J Interact Des Manuf 17, 899–916 (2023). https://doi.org/10.1007/s12008-022-01175-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12008-022-01175-7