DOI QR코드

DOI QR Code

Machine learning-enabled parameterization scheme for aerodynamic shape optimization of wind-sensitive structures: A-proof-of-concept study

  • Shaopeng Li (University of Louisiana at Lafayette) ;
  • Brian M. Phillips (University of Florida) ;
  • Zhaoshuo Jiang (San Francisco State University)
  • Received : 2024.02.16
  • Accepted : 2024.08.06
  • Published : 2024.09.25

Abstract

Aerodynamic shape optimization is very useful for enhancing the performance of wind-sensitive structures. However, shape parameterization, as the first step in the pipeline of aerodynamic shape optimization, still heavily depends on empirical judgment. If not done properly, the resulting small design space may fail to cover many promising shapes, and hence hinder realizing the full potential of aerodynamic shape optimization. To this end, developing a novel shape parameterization scheme that can reflect real-world complexities while being simple enough for the subsequent optimization process is important. This study proposes a machine learning-based scheme that can automatically learn a low-dimensional latent representation of complex aerodynamic shapes for bluff-body wind-sensitive structures. The resulting latent representation (as design variables for aerodynamic shape optimization) is composed of both discrete and continuous variables, which are embedded in a hierarchy structure. In addition to being intuitive and interpretable, the mixed discrete and continuous variables with the hierarchy structure allow stakeholders to narrow the search space selectively based on their interests. As a proof-of-concept study, shape parameterization examples of tall building cross sections are used to demonstrate the promising features of the proposed scheme and guide future investigations on data-driven parameterization for aerodynamic shape optimization of wind-sensitive structures.

Keywords

Acknowledgement

This material is based upon work supported by the National Science Foundation (NSF) under Grants No. 2028762 & 2028647. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of NSF.

References

  1. Abbas, T., Kavrakov, I., Morgenthal, G. and Lahmer, T., (2022), "Framework for a simulation-based aerodynamic shape optimization of bridge decks for different limit state phenomena", arXiv preprint arXiv:2203.14414.
  2. Alkhatib, F., Kasim, N., Goh, W.I., Shafiq, N., Amran, M., Kotov, E.V. and Albaom, M.A., (2022), "Computational aerodynamic optimization of wind-sensitive irregular tall buildings", Buildings, 12(7), 939. https://doi.org/10.3390/buildings12070939.
  3. Allen, C.B., Poole, D.J. and Rendall, T. (2018), "Wing aerodynamic optimization using efficient mathematically-extracted modal design variables", Optimiz. Eng., 19(2), 453-477. https://doi.org/10.1007/s11081-018-9376-7.
  4. Barelli, M., White, J. and Billington, D.P., (2006), "History and aesthetics of the Bronx-Whitestone Bridge", J. Bridge Eng., 11(2), 230-240. https://doi.org/10.1061/(ASCE)1084-0702(2006)11:2(230.
  5. Bernardini, E., Spence, S.M., Wei, D. and Kareem, A. (2015), "Aerodynamic shape optimization of civil structures: A CFD-enabled Kriging-based approach", J. Wind Eng. Ind. Aerod., 144, 154-164. https://doi.org/10.1016/j.jweia.2015.03.011.
  6. Birhane, T.H., Bitsuamlak, G.T. and King, J.P., (2017), "A computational framework for the aerodynamic shape optimization of long-span bridge decks", In Structures Congress 2017, 223-239.
  7. Chen, W., Chiu, K. and Fuge, M.D. (2020), "Airfoil design parameterization and optimization using bezier generative adversarial networks", AIAA J., 58(11), 4723-4735. https://doi.org/10.2514/1.J059317.
  8. Chen, W., Fuge, M. and Chazan, J. (2017), "Design manifolds capture the intrinsic complexity and dimension of design spaces", J. Mech. Des., 139(5). https://doi.org/10.1115/1.4036134.
  9. Cinquegrana, D. and Iuliano, E., (2017), "Efficient global optimization of a transonic wing with geometric data reduction", In 35th AIAA Applied Aerodynamics Conference.
  10. Davenport, A.G. (1971), "The response of six building shapes to turbulent wind", Philosoph. Transact. Royal Soc. London. Series A, Mathem. Phys. Sci., 269(1199), 385-394. https://doi.org/10.1098/rsta.1971.0039.
  11. Ding, F. and Kareem, A. (2018), "A multi-fidelity shape optimization via surrogate modeling for civil structures", J. Wind Eng. Ind. Aerod., 178, 49-56. https://doi.org/10.1016/j.jweia.2018.04.022.
  12. Dulikravich, G.S. (1992), "Aerodynamic shape design and optimization-status and trends", J. Aircraft, 29(6), 1020-1026. https://doi.org/10.2514/3.46279.
  13. Elshaer, A. and Bitsuamlak, G. (2018), "Multiobjective aerodynamic optimization of tall building openings for wind-induced load reduction", J. Struct. Eng., 144(10), 04018198. https://doi.org/10.1061/(ASCE)ST.1943-541X.0002199.
  14. Elshaer, A., Bitsuamlak, G. and El Damatty, A. (2016), "Aerodynamic shape optimization of tall buildings using twisting and corner modifications", In 8th International Colloquium on Bluff Body Aerodynamics and Applications.
  15. Elshaer, A., Bitsuamlak, G. and El Damatty, A. (2017), "Enhancing wind performance of tall buildings using corner aerodynamic optimization", Eng. Struct., 136, 133-148. https://doi.org/10.1016/j.engstruct.2017.01.019.
  16. Fujii, K., Saito, Y., Takamichi, S., Baba, Y. and Saruwatari, H. (2020), "HumanGAN: generative adversarial network with human-based discriminator and its evaluation in speech perception modeling", In Proceedings of 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.
  17. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. and Bengio, Y. (2014), "Generative adversarial nets", Adv. Neural Inform. Process. Syst., 27.
  18. Google Earth (2023), https://earth.google.com/, Accessed on July 24, 2023.
  19. Hallaji, E., Farajzadeh-Zanjani, M., Razavi-Far, R., Palade, V. and Saif, M., (2021), "Constrained generative adversarial learning for dimensionality reduction", IEEE Transact. Knowledge Data Eng., https://doi.org/10.1109/TKDE.2021.3126642.
  20. He, Z., Liang, T., Lai, X., Gao, M., Tu, X. and Lu, Y., (2022), "Vibration acceleration-integrated parameterized aerodynamic shape optimization of super high-rise buildings with spiral configurations", Struct. Multidiscipl. Optimiz., 65(10), 1-21. https://doi.org/10.1007/s00158-022-03387-7.
  21. Hinton, G.E. and Salakhutdinov, R.R. (2006), "Reducing the dimensionality of data with neural networks", Science, 313(5786), 504-507. https://doi.org/10.1126/science.1127647.
  22. Jafari, M. and Alipour, A., (2021), "Review of approaches, opportunities, and future directions for improving aerodynamics of tall buildings with smart facades", Sustain. Cities Soci., 72, 102979. https://doi.org/10.1016/j.scs.2021.102979.
  23. Jaouadi, Z., Abbas, T., Morgenthal, G. and Lahmer, T. (2020), "Single and multi-objective shape optimization of streamlined bridge decks", Struct. Multidiscipl. Optimiz., 61(4),1495-1514. https://doi.org/10.1007/s00158-019-02431-3.
  24. Kim, H. and Mnih, A. (2018), "Disentangling by factorizing", In International Conference on Machine Learning. PMLR.
  25. Larsen, A. (1993), "Aerodynamic aspects of the final design of the 1624 m suspension bridge across the Great Belt", J. Wind Eng. Ind. Aerod., 48(2-3), 261-285. https://doi.org/10.1016/0167-6105(93)90141-A.
  26. Larsen, A., Esdahl, S., Andersen, J.E. and Vejrum, T. (2000), "Storebaelt suspension bridge-vortex shedding excitation and mitigation by guide vanes", J. Wind Eng. Ind. Aerod., 88(2-3), 283-296. https://doi.org/10.1016/S0167-6105(00)00054-4.
  27. Lepine, J., Guibault, F., Trepanier, J.Y. and Pepin, F. (2001), "Optimized nonuniform rational B-spline geometrical representation for aerodynamic design of wings", AIAA J., 39(11), 2033-2041.
  28. Li, S., Snaiki, R. and Wu, T. (2021), "A knowledge-enhanced deep reinforcement learning-based shape optimizer for aerodynamic mitigation of wind-sensitive structures", Comput. Aided Civil Infrastruct. Eng., 36(6), 733-746. https://doi.org/10.1111/mice.12655.
  29. Li, X., Lin, C., Li, R., Wang, C. and Guerin, F. (2020), "Latent space factorisation and manipulation via matrix subspace projection", In Proceedings of 37th International Conference on Machine Learning.
  30. Montoya, M.C., Hernandez, S. and Nieto, F., (2018), "Shape optimization of streamlined decks of cable-stayed bridges considering aeroelastic and structural constraints", J. Wind Eng. Ind. Aerod., 177, 429-455. https://doi.org/10.1016/j.jweia.2017.12.018.
  31. Mooneghi, M.A. and Kargarmoakhar, R. (2016), "Aerodynamic mitigation and shape optimization of buildings", J. Build. Eng., 6, 225-235. https://doi.org/10.1016/j.jobe.2016.01.009.
  32. Nagao, F., Utsunomiya, H., Oryu, T. and Manabe, S. (1993), "Aerodynamic efficiency of triangular fairing on box girder bridge", J. Wind Eng. Ind. Aerod., 49(1-3), 565-574. https://doi.org/10.1016/0167-6105(93)90050-X.
  33. Nieto, F., Cid Montoya, M. and Hernandez, S. (2022), "Shape optimization of tall buildings cross-section: Balancing profit and aeroelastic performance", Struct. Des. Tall Spec. Build., e1982. https://doi.org/10.1002/tal.1982.
  34. Nieto, F., Montoya, M.C., Hernandez, S., Kusano, I., Casteleiro, A., Alvarez, A.J., Jurado, J.A. and Fontan, A. (2020), "Aerodynamic and aeroelastic responses of short gap twin-box decks: Box geometry and gap distance dependent surrogate based design", J. Wind Eng. Ind. Aerod., 201, 104147. https://doi.org/10.1016/j.jweia.2020.104147.
  35. Omohundro, S.M. (1989), Five Balltree Construction Algorithms. International Computer Science Institute, Berkeley.
  36. RealCity3D (2023), A Large-scale Georeferenced 3D Shape Dataset of Real-world Cities, https://ai4ce.github.io/RealCity3D/, accessed on July 24, 2023.
  37. Ross, A. and Doshi-Velez, F. (2021), "Benchmarks, algorithms, and metrics for hierarchical disentanglement", In Proceedings of the 38th International Conference on Machine Learning, PMLR, 139, 2021.
  38. Ross, A., Chen, N., Hang, E. Z., Glassman, E.L. and Doshi-Velez, F. (2021), "Evaluating the interpretability of generative models by interactive reconstruction", In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems.
  39. Sakai, Y., Ogawa, K., Shimodoi, H. and Saitoh, T. (1993), "An experimental study on aerodynamic improvements for edge girder bridges", J. Wind Eng. Ind. Aerod., 49(1-3), 459-466. https://doi.org/10.1016/0167-6105(93)90040-U.
  40. Sharma, A., Mittal, H. and Gairola, A. (2018), "Mitigation of wind load on tall buildings through aerodynamic modifications", J. Build. Eng., 18, 180-194. https://doi.org/10.1016/j.jobe.2018.03.005.
  41. Shirzadi, M. and Tominaga, Y. (2021), "Multi-fidelity shape optimization methodology for pedestrian-level wind environment", Build. Environ., 204, 108076. https://doi.org/10.1016/j.buildenv.2021.108076.
  42. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M. and Kim, Y.C. (2012), "Experimental investigation of aerodynamic forces and wind pressures acting on tall buildings with various unconventional configurations", J. Wind Eng. Ind. Aerod., 107, 179-191. https://doi.org/10.1016/j.jweia.2012.04.014.
  43. Tang, H., Li, Y., Wang, Y. and Tao, Q. (2017), "Aerodynamic optimization for flutter performance of steel truss stiffening girder at large angles of attack", J. Wind Eng. Ind. Aerod., 168, 260-270. https://doi.org/10.1016/j.jweia.2017.06.013.
  44. Tang, J.W., Xie, Y.M., Felicetti, P., Tu, J.Y. and Li, J.D. (2013), "Numerical simulations of wind drags on straight and twisted polygonal buildings", Struct. Des. Tall Spec. Build., 22(1), 62-73. https://doi.org/10.1002/tal.657.
  45. Tinmitonde, S., He, X. and Yan, L. (2022), "Single-objective aerodynamic optimization of a streamlined bridge deck subjected to shape modification using a polynomial emulator and genetic algorithm", Struct. Multidiscipl. Optimiz., 65(12), 1-21. https://doi.org/10.1007/s00158-022-03459-8.
  46. Topping, B.H.V. (1983), "Shape optimization of skeletal structures: a review", J. Struct. Eng., 109(8), 1933-1951. https://doi.org/10.1061/(ASCE)0733-9445(1983)109:8(1933).
  47. UIUC Airfoil Coordinates Database, https://mselig.ae.illinois.edu/ads/coord_database.html.
  48. Venkataraman, P. (1995), "A new procedure for airfoil definition", In 13th Applied Aerodynamics Conference.
  49. Wang, Y., Yao, H. and Zhao, S. (2016), "Auto-encoder based dimensionality reduction", Neurocomputing, 184, 232-242. https://doi.org/10.1016/j.neucom.2015.08.104.
  50. Wang, Z., Zheng, C., Mulyanto, J.A. and Wu, Y. (2022), "Aerodynamic shape optimization of a square cylinder with multi-parameter corner recession modifications", Atmosphere, 13(11), 1782. https://doi.org/10.3390/atmos13111782.
  51. Whiteman, M.L., Fernandez-Caban, P.L., Phillips, B.M., Masters, F.J., Davis, J.R. and Bridge, J.A. (2022), "Cyber-physical aerodynamic shape optimization of a tall building in a wind tunnel using an active fin system", J. Wind Eng. Ind. Aerod., 220, 104835. https://doi.org/10.1016/j.jweia.2021.104835.
  52. Yang, Y., Wu, T., Ge, Y. and Kareem, A. (2015), "Aerodynamic stabilization mechanism of a twin box girder with various slot widths", J. Bridge Eng., 20(3), 04014067. https://doi.org/10.1061/(ASCE)BE.1943-5592.0000645.