Structural Engineering and Mechanics

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Form-finding of lifting self-forming GFRP elastic gridshells based on machine learning interpretability methods

  • Soheila, Kookalani (Department of Civil Engineering, Shanghai Jiao Tong University) ;
  • Sandy, Nyunn (Department of Civil Engineering, Shanghai Jiao Tong University) ;
  • Sheng, Xiang (Department of Civil Engineering, Shanghai Jiao Tong University)
  • Received : 2022.04.15
  • Accepted : 2022.11.04
  • Published : 2022.12.10
⟨ Previous Next ⟩

Abstract

Glass fiber reinforced polymer (GFRP) elastic gridshells consist of long continuous GFRP tubes that form elastic deformations. In this paper, a method for the form-finding of gridshell structures is presented based on the interpretable machine learning (ML) approaches. A comparative study is conducted on several ML algorithms, including support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), AdaBoost, XGBoost, category boosting (CatBoost), and light gradient boosting machine (LightGBM). A numerical example is presented using a standard double-hump gridshell considering two characteristics of deformation as objective functions. The combination of the grid search approach and k-fold cross-validation (CV) is implemented for fine-tuning the parameters of ML models. The results of the comparative study indicate that the LightGBM model presents the highest prediction accuracy. Finally, interpretable ML approaches, including Shapely additive explanations (SHAP), partial dependence plot (PDP), and accumulated local effects (ALE), are applied to explain the predictions of the ML model since it is essential to understand the effect of various values of input parameters on objective functions. As a result of interpretability approaches, an optimum gridshell structure is obtained and new opportunities are verified for form-finding investigation of GFRP elastic gridshells during lifting construction.

Keywords

Acknowledgement

The research work was supported by the National Natural Science Foundation of China (project No. 51978400) and the National Key Research and Development Program of China (project No. 2021YFE0107800). The support is gratefully acknowledged.

References

  1. Apley, D.W. and Zhu, J. (2020), "Visualizing the effects of predictor variables in black box supervised learning models", J. Roy. Stat. Soc.. Ser. B: Stat. Meth., 82(4), 1059-1086. https://doi.org/10.1111/rssb.12377.
  2. Baek, C. and Reis, P.M. (2019), "Rigidity of hemispherical elastic gridshells under point load indentation", J. Mech. Phys. Solid., 124, 411-426. https://doi.org/10.1016/j.jmps.2018.11.002.
  3. Belle, V. and Papantonis, I. (2021), "Principles and practice of explainable machine learning", Front. Big Data, https://doi.org/10.3389/fdata.2021.688969.
  4. Bouhaya, L., Baverel, O. and Caron, J.F. (2009), "Mapping twoway continuous elastic grid on an imposed surface: Application to grid shells", Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2009, Valencia.
  5. Breiman, L. (2001), "Random forests", Mach. Learn., 45(1), 5-32. https://doi.org/10.1023/A:1010933404324.
  6. Carbonell-Marquez, J.F., Gil-Martin, L.M., Fernandez-Ruiz, M.A. and Hernandez-Montes, E. (2016), "Topological design of compression structures", Arch. Appl Mech., 86(8), 1495-1508. https://doi.org/10.1007/s00419-016-1131-y.
  7. Chen, T., He, T., Benesty, M., Khotilovich, V., Tang, Y., Cho, H. and Chen, K. (2015), "Xgboost: extreme gradient boosting", R Package Version 0.4-2, 1(4), 1-4.
  8. Chen, T. and Guestrin, C. (2016), "XGBoost: A scalable tree boosting system", Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794. https://doi.org/10.1145/2939672.2939785.
  9. Chopra, P., Sharma, R.K., Kumar, M. and Chopra, T. (2018), "Comparison of machine learning techniques for the prediction of compressive strength of concrete", Adv. Civil Eng., 2018, Article ID 5481705. https://doi.org/10.1155/2018/5481705.
  10. Cover, T.M. and Hart, P.E. (1967), "Nearest neighbor pattern classification", IEEE Tran. Inform. Theory, 13(1), 21-27. https://doi.org/10.1109/TIT.1967.1053964.
  11. Dang, H.V., Raza, M., Tran-Ngoc, H., Bui-Tien, T. and Nguyen, H.X. (2021), "Connection stiffness reduction analysis in steel bridge via deep CNN and modal experimental data", Struct. Eng. Mech., 77(4), 495-508. https://doi.org/10.12989/sem.2021.77.4.495.
  12. Das, S., Dutta, S., Putcha, C., Majumdar, S. and Adak, D. (2020), "A Data-driven physics-informed method for prognosis of infrastructure systems: Theory and application to crack prediction", ASCE-ASME J. Risk Uncert. Eng. Syst., Part A: Civil Eng., 6(2), 04020013. https://doi.org/10.1061/ajrua6.0001053.
  13. Dietterich, T.G. (2000), "Ensemble methods in machine learning", Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).
  14. Dorogush Veronika, A., Ershov, V. and Gulin, A. (2018), "CatBoost: gradient boosting with categorical features support", arXiv preprint arXiv:1810.11363.
  15. Fernandes, J.G., Kirkegaard, P.H. and Branco, J.M. (2016), "Tectonic design of elastic timber gridshells", WCTE 2016-World Conference on Timber Engineering.
  16. Fernandez-Ruiz, M.A., Hernandez-Montes, E., CarbonellMarquez, J.F. and Gil-Martin, L.M. (2017), "Patterns of force:length ratios for the design of compression structures with inner ribs", Eng. Struct., 148, 878-889. https://doi.org/10.1016/j.engstruct.2017.07.027.
  17. Friedman, J.H. (2001), "Greedy function approximation: A gradient boosting machine", Ann. Stat., 1189-1232. https://doi.org/10.1214/aos/1013203451.
  18. Hernandez-Montes, E., Jurado-Pina, R. and Bayo, E. (2006), "Topological mapping for tension structures", J. Struct. Eng., 132(6), 970-977. https://doi.org/10.1061/(asce)0733-9445(2006)132:6(970).
  19. Hernandez, E.L., Baverel, O. and Gengnagel, C. (2013), "On the design and construction of elastic gridshells with irregular meshes", Int. J. Space Struct., 28(3-4), 161-174. https://doi.org/10.1260/0266-3511.28.3-4.161.
  20. Huang, H. and Burton, H.V. (2019), "Classification of in-plane failure modes for reinforced concrete frames with infills using machine learning", J. Build. Eng., 25, 100767. https://doi.org/10.1016/j.jobe.2019.100767.
  21. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T.Y. (2017), "LightGBM: A highly efficient gradient boosting decision tree", Advances in Neural Information Processing Systems, 30.
  22. Kim-Lan Vaulot, L. (2016), "Form-finding of elastic gridshells", PhD Thesis, Massachusetts Institute of Technology.
  23. Kookalani, S., Cheng, B. and Xiang, S. (2021), "Shape optimization of GFRP elastic gridshells by the weighted Lagrange ε-twin support vector machine and multi-objective particle swarm optimization algorithm considering structural weight", Struct., 33, 2066-2084. https://doi.org/10.1016/j.istruc.2021.05.077.
  24. Kopper, P. (2019), "Limitations of interpretable machine learning methods".
  25. Lafuente Hernandez, E., Sechelmann, S., Rorig, T. and Gengnagel, C. (2013), "Topology optimisation of regular and irregular elastic gridshells by means of a non-linear variational method", Adv. Arch. Geometry, 2012, 147-160. https://doi.org/10.1007/978-3-7091-1251-9_11.
  26. Lamberti, L. (2008), "An efficient simulated annealing algorithm for design optimization of truss structures", Comput. Struct., 86(19-20), 1936-1953. https://doi.org/10.1016/j.compstruc.2008.02.004.
  27. Lundberg, S.M. and Lee, S.I. (2017), "A unified approach to interpreting model predictions", Advances in Neural Information Processing Systems, 30.
  28. Mangalathu, S., Jang, H., Hwang, S.H. and Jeon, J.S. (2020), "Data-driven machine-learning-based seismic failure mode identification of reinforced concrete shear walls", Eng. Struct., 208, 110331. https://doi.org/10.1016/j.engstruct.2020.110331.
  29. Mangalathu, S. and Jeon, J.S. (2018), "Classification of failure mode and prediction of shear strength for reinforced concrete beam-column joints using machine learning techniques", Eng. Struct., 160, 85-94. https://doi.org/10.1016/j.engstruct.2018.01.008.
  30. Mao, J.X., Wang, H., Feng, D.M., Tao, T.Y. and Zheng, W.Z. (2018), "Investigation of dynamic properties of long-span cablestayed bridges based on one-year monitoring data under normal operating condition", Struct. Control Hlth. Monit., 25(5), e2146. https://doi.org/10.1002/stc.2146.
  31. Mesnil, R., Ochsendorf, J. and Douthe, C. (2015), "Stability of pseudo-funicular elastic grid shells", Int. J. Space Struct., 30(1), 27-36. https://doi.org/10.1260/0266-3511.30.1.27.
  32. Molnar, C. (2019), Interpretable Machine Learning. A Guide for Making Black Box Models Explainable.
  33. Murdoch, W.J., Singh, C., Kumbier, K., Abbasi-Asl, R. and Yu, B. (2019), "Definitions, methods, and applications in interpretable machine learning", Proc. Nat. Acad. Sci. US Am., 116(44), 22071-22080 https://doi.org/10.1073/pnas.1900654116.
  34. Nguyen, H.D., Bui Tien, T., De Roeck, G. and Abdel Wahab, M. (2021), "Damage detection in structures using modal curvatures gapped smoothing method and deep learning", Struct. Eng. Mech., 77(1), 47-56. https://doi.org/10.12989/sem.2021.77.1.047.
  35. Omidi, M., Liu, S.J., Mohtaram, S., Lu, H.T. and Zhang, H.C. (2020), "Flow simulation and model analysis of efficiency and pressure ratio behaviours in GT4086 turbocharger compressor", Mech., 26(4), 293-300. https://doi.org/10.5755/j01.mech.26.4.23867.
  36. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., ... & Duchesnay, E. (2011), "Scikit-learn: Machine learning in {P}ython", J. Mach. Learn. Res., 12, 2825-2830.
  37. Du Peloux, L., Baverel, O., Caron, J.F. and Tayeb, F. (2011), "From shape to shell: A design tool to materialize freeform shapes using gridshell structures", Design Modeling Symposium. Berlin, September.
  38. Saka, M.P. (2007), "Optimum topological design of geometrically nonlinear single layer latticed domes using coupled genetic algorithm", Comput. Struct., 85(21-22), 1635-1646. https://doi.org/10.1016/j.compstruc.2007.02.023.
  39. Slater, E., Moni, M. and Alam, M.S. (2012), "Predicting the shear strength of steel fiber reinforced concrete beams", Constr. Build. Mater., 26(1), 423-436. https://doi.org/10.1016/j.conbuildmat.2011.06.042.
  40. Tayeb, F., Caron, J.F., Baverel, O. and Du Peloux, L. (2013), "Stability and robustness of a 300 m2 composite gridshell structure", Constr. Build. Mater., 49, 926-938. https://doi.org/10.1016/j.conbuildmat.2013.04.036.
  41. Safa, M. and Kachitvichyanukul, V. (2019), "Moment-rotation prediction of precast beam-to-column connections using extreme learning machine", Struct. Eng. Mech., 70(5), 639-647. https://doi.org/10.12989/sem.2019.70.5.639.
  42. Wang, H., Xu, Z., Wu, T. and Mao, J. (2018), "Evolutionary power spectral density of recorded typhoons at Sutong Bridge using harmonic wavelets", J. Wind Eng. Indus. Aerodyn., 177, 197-212. https://doi.org/10.1016/j.jweia.2018.04.015.
  43. Xiang, S., Cheng, B., Kookalani, S. and Zhao, J. (2021), "An analytic approach to predict the shape and internal forces of barrel vault elastic gridshells during lifting construction", Struct., 29, 628-637. https://doi.org/10.1016/j.istruc.2020.11.032.
  44. Xiang, S., Cheng, B. and Kookalani, S. (2020), "An analytic solution for form finding of GFRP elastic gridshells during lifting construction", Compos. Struct., 244, 112290. https://doi.org/10.1016/j.compstruct.2020.112290.
  45. Zhang, C. and Ma, Y. (2012), Ensemble machine learning: Methods and applications, Ensemble Machine Learning: Methods and Applications, Springer Science & Business Media.