Wind and Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

LSTM-based aerodynamic force modeling for unsteady flows around structures

  • Shijie Liu (School of Civil Engineering, Shijiazhuang Tiedao University) ;
  • Zhen Zhang (School of Civil Engineering, Shijiazhuang Tiedao University) ;
  • Xue Zhou (School of Civil Engineering, Shijiazhuang Tiedao University) ;
  • Qingkuan Liu (School of Civil Engineering, Shijiazhuang Tiedao University)
  • Received : 2023.06.05
  • Accepted : 2024.01.22
  • Published : 2024.02.25
⟨ Previous Next ⟩

Abstract

The aerodynamic force is a significant component that influences the stability and safety of structures. It has unstable properties and depends on computer precision, making its long-term prediction challenging. Accurately estimating the aerodynamic traits of structures is critical for structural design and vibration control. This paper establishes an unsteady aerodynamic time series prediction model using Long Short-Term Memory (LSTM) network. The unsteady aerodynamic force under varied Reynolds number and angles of attack is predicted by the LSTM model. The input of the model is the aerodynamic coefficients of the 1 to n sample points and output is the aerodynamic coefficients of the n+1 sample point. The model is predicted by interpolation and extrapolation utilizing Unsteady Reynolds-average Navier-Stokes (URANS) simulation data of flow around a circular cylinder, square cylinder and airfoil. The results illustrate that the trajectories of the LSTM prediction results and URANS outcomes are largely consistent with time. The mean relative error between the forecast results and the original results is less than 6%. Therefore, our technique has a prospective application in unsteady aerodynamic force prediction of structures and can give technical assistance for engineering applications.

Keywords

References

  1. Achenbach, E. (1968), "Distribution of local pressure and skin friction around a circular cylinder in cross-flow up to Re = 5 × 106", J. Fluid Mech., 34(4), 625-639. https://doi.org/10.1017/S0022112068002120.
  2. Bengio, Y., Simard, P. and Frasconi, P. (1994). "Learning long-term dependencies with gradient descent is difficult", IEEE Trans. Neural Netw., 5(2), 157-166. https://doi.org/10.1109/72.279181.
  3. Bertani, G., Patruno, L. and Aguera, F.G. (2022), "Low-fidelity simulations in Computational Wind Engineering: shortcomings of 2D RANS in fully separated flows", Wind Struct., 34, 499-510. https://doi.org/10.12989/WAS.2022.34.6.499.
  4. Blocken, B. (2014), "50 years of Computational Wind Engineering: Past, present and future", J. Wind Eng. Ind. Aerody., 129, 69-102. https://doi.org/10.1016/j.jweia.2014.03.008.
  5. Carassale, L., Solari, G. and Tubino, F. (2007), "Proper orthogonal decomposition in wind engineering - Part 2: Theoretical aspects and some applications", Wind Struct., 10(2), 177-208. https://doi.org/10.12989/WAS.2007.10.2.177.
  6. Cheng, L., Lam, K.M. and Wong, S.Y. (2015), "POD analysis of crosswind forces on a tall building with square and H-shaped cross sections", Wind Struct., 21(1), 63-84. https://doi.org/10.12989/WAS.2015.21.1.063.
  7. Gao, D.L., Chen, W.L., Li, H. and Hu, H. (2017), "Flow around a slotted circular cylinder at various angles of attack", Exp Fluids, 58, 132. https://doi.org/10.1007/s00348-017-2417-8.
  8. Huang, D.M., He, S.Q., He, X.H. and Zhu, X. (2017), "Prediction of wind loads on high-rise building using a BP neural network combined with POD", J. Wind Eng. Ind. Aerod., 170, 1-17. https://doi.org/10.1016/j.jweia.2017.07.021.
  9. Hochreiter, S. and Schmidhuber, J. (1997), "Long short-term memory", Neural Comput., 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735.
  10. Hu, G. and Kwok, K.C.S. (2020), "Predicting wind pressures around circular cylinders using machine learning techniques", J. Wind Eng. Ind. Aerod., 198, 104099. https://doi.org/10.1016/j.jweia.2020.104099.
  11. Huang, D., Wu, T. and He, S. (2020), "Experimental investigation of vortex-induced aeroelastic effects on a square cylinder in uniform flow", Wind Struct., 30, 37-54. https://doi.org/10.12989/WAS.2020.30.1.037.
  12. Jiang, H.Y. and Cheng, L. (2021), "Large-eddy simulation of flow past a circular cylinder for Reynolds numbers 400 to 3900", Phys. Fluids, 33(3), 034119. https://doi.org/10.1063/5.0041168.
  13. Jin, X.W., Cheng, P., Chen, W.L. and Li, H. (2018), "Prediction model of velocity field around circular cylinder over various Reynolds numbers by fusion convolutional neural networks based on pressure on the cylinder", Phys. Fluids, 30(4), 047105. https://doi.org/10.1063/1.5024595.
  14. Jing, H.Q., He, X.H. and Li H. (2018), "Initial condition and damping effects on rain-wind induced cable vibration", J. Wind Eng. Ind. Aerod., 175, 376-383. https://doi.org/10.1016/j.jweia.2018.02.011.
  15. Kim, D.H. and Chang, J.W. (2014), "Low-Reynolds-number effect on the aerodynamic characteristics of a pitching NACA0012 airfoil", Aerosp. Sci. Technol., 32(1), 162-168. https://doi.org/10.1016/j.ast.2013.08.018.
  16. Kozmar, H., Procino, L., Borsani, A. and Bartoli, G. (2012), "Sheltering efficiency of wind barriers on bridges", J. Wind Eng. Ind. Aerod., 107-108, 274-284. https://doi.org/10.1016/j.jweia.2012.04.027.
  17. Kutz, J.N. (2017), "Deep learning in fluid dynamics", J. Fluid Mech., 814, 1-4. https://doi.org/10.1017/jfm.2016.803.
  18. Kwok, K.C.S., Hitchcock, P.A., Burton, M.D. (2009), "Perception of vibration and occupant comfort in wind-excited tall buildings", J. Wind Eng. Ind. Aerod., 97(7-8), 368-380. https://doi.org/10.1016/j.jweia.2009.05.006.
  19. Laima, S.J., Feng, H., Li, H., Jin, Y., Han, F.Y. and Xu, W.C. (2023), "A Buffeting-Net for buffeting response prediction of full-scale bridges", Eng. Struct., 275, 115289. https://doi.org/10.1016/j.engstruct.2022.115289.
  20. Laima, S.J., Jiang, C., Li, H., Chen, W.L. and Ou, J.P. (2018), "A numerical investigation of Reynolds number sensitivity of flow characteristics around a twin-box girder", J. Wind Eng. Ind. Aerod., 172, 298-316. https://doi.org/10.1016/j.jweia.2017.11.016.
  21. Lecun, Y., Bengio, Y. and Hinton, G. (2015), "Deep learning", Nature, 521(7553), 436-444. https://doi.org/10.1038/nature14539.
  22. Li, H., Laima, S.J. and Jing, H.Q. (2014), "Reynolds number effects on aerodynamic characteristics and vortex-induced vibration of a twin-box girder", J. Fluids Struct., 50, 358-375. https://doi.org/10.1016/j.jfluidstructs.2014.06.027.
  23. Li, S., Laima, S. and Li, H. (2017), "Cluster analysis of winds and wind-induced vibrations on a long-span bridge based on long-term field monitoring data", Eng. Struct., 138, 245-259. https://doi.org/10.1016/j.engstruct.2017.02.024.
  24. Li, T., Wu, T. and Liu, Z. (2020), "Nonlinear unsteady bridge aerodynamics: Reduced-order modeling based on deep LSTM networks", J. Wind Eng. Ind. Aerod., 198, 104116. https://doi.org/10.1016/j.jweia.2020.104116.
  25. Li, W.J., Laima, S.J., Jin, X.W., Yuan, W.Y. and Li, H. (2020), "A novel long short-term memory neural-network-based self-excited force model of limit cycle oscillations of nonlinear flutter for various aerodynamic configurations", Nonlinear Dynam., 100(3), 2071-2087. https://doi.org/10.1007/s11071-020-05631-5.
  26. Lienard, C., Boisard, R. and Daudin, C. (2019) "Aerodynamic behavior of a floating offshore wind turbine", AIAA Scitech 2019 Forum, San Diego, California, January.
  27. Ma, W.Y., Liu, Q.K., Macdonald, J.H.G. and Yan, X.D., Zheng, Y.F. (2019), "The effect of surface roughness on aerodynamic forces and vibrations for a circular cylinder in the critical Reynolds number range", J. Wind Eng. Ind. Aerod., 187, 61-72. https://doi.org/10.1016/j.jweia.2019.01.011.
  28. Niemann, H.J. and Holscher, N. (1990), "A review of recent experiments on the flow past circular cylinders", J. Wind Eng. Ind. Aerodyn., 33(1-2), 197-209. https://doi.org/10.1016/0167-6105(90)90035-B.
  29. Norberg, C. (1993), "Flow around rectangular cylinders: Pressure forces and wake frequencies", J. Wind Eng. Ind. Aerod., 49, 187-196. https://doi.org/10.1016/0167-6105(93)90014-F.
  30. Pathak, J., Hunt, B., Girvan, M., Lu, Z.X. and Ott, E. (2018), "Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach", Phys. Rev. Lett., 120(2), 024102. https://doi.org/10.1103/PhysRevLett.120.024102.
  31. Rumelhart, D.E., Hinton, G.E. and Williams, R.J. (1986), "Learning representations by back-propagating errors", Nature, 323(6088), 533-536. https://doi.org/10.1038/323533a0.
  32. Ryu, S., Kim, S., Choi, J., Yu, H. and Lee, G.G. (2017), "Neural Sentence Embedding using Only In-domain Sentences for Out-of-domain Sentence Detection in Dialog Systems", Pattern Recognition Letters, 88, 26-32. https://doi.org/10.1016/j.patrec.2017.01.008.
  33. Shi G.T. and Li Z.P. (2013), "Effect of reynolds number on low speed symmetrical airfoil aerodynamic performance", New Technology and New Process, 9, 18-22.
  34. Stringer, R.M., Zang, J. and Hillis, A.J. (2014), "Unsteady RANS computations of flow around a circular cylinder for a wide range of Reynolds numbers", Ocean Eng., 87, 1-9. https://doi.org/10.1016/j.oceaneng.2014.04.017.
  35. Sun, Y.F., Liu, Q.K., Shao, L.Y., Wang, Y.X., Chang, X. and Liu, K. (2023), "Experimental investigation of aerodynamic forces and vortex-induced vibrations of wavy cylinders at subcritical Reynolds numbers", Exp. Therm. Fluid Sci., 144, 110869. https://doi.org/10.1016/j.expthermflusci.2023.110869.
  36. Szechenyi, E. (1975), "Supercritical Reynolds number simulation for two-dimensional flow over circular cylinders", J. Fluid Mech., 70(3), 529-542. https://doi.org/10.1017/S0022112075002170.
  37. Thom, A. (1933), "The flow past circular cylinders at low speeds", Proc. R. Soc. Lond. A, 141(845), 651-669. https://doi.org/10.1098/rspa.1933.0146.
  38. Tamura, T. and Miyagi, T. (1999), "The effect of turbulence on aerodynamic forces on a square cylinder with various corner shapes", J. Wind Eng. Ind. Aerodyn., 83, 135-145. https://doi.org/10.1016/S0167-6105(99)00067-7.
  39. Wang, H.F. and Wu, T. (2020), "Knowledge-Enhanced Deep Learning for Wind-Induced Nonlinear Structural Dynamic Analysis", J. Struct. Eng., 146(11), 04020235. https://doi.org/10.1061/(ASCE)ST.1943-541X.0002802.
  40. Wang, H., Zhang, Y.M., Mao, J.X. and Wan, H.P. (2020), "A probabilistic approach for short-term prediction of wind gust speed using ensemble learning", J. Wind Eng. Ind. Aerod., 202, 104198. https://doi.org/10.1016/j.jweia.2020.104198.
  41. Wu, T. and Snaiki, R. (2022), "Applications of Machine Learning to Wind Engineering", Front. Built Environ., 8, 811460. https://doi.org/10.3389/fbuil.2022.811460.
  42. Yagmur, S., Dogan, S., Aksoy, M.H., Canli, E. and Ozgoren, M. (2015), "Experimental and Numerical Investigation of Flow Structures around Cylindrical Bluff Bodies", Epj Web Conf., 92, 02113. https://doi.org/10.1051/epjconf/20159202113.
  43. Yang, J., Guo, Y.Q. and Zhao, W.L. (2019), "Long short-term memory neural network based fault detection and isolation for electro-mechanical actuators", Neurocomputing, 360, 85-96. https://doi.org/10.1016/j.neucom.2019.06.029.
  44. Zhai, G.H., Narazaki, Y., Wang, S., Shajihan, S.A.V. and Spencer, B.F.Jr. (2022), "Synthetic data augmentation for pixel-wise steel fatigue crack identification using fully convolutional networks", Smart Struct. Syst., 29(1), 237-250. https://doi.org/10.12989/SSS.2022.29.1.237.
  45. Zhang, M., Zhao, G.F. and Wang, J.L. (2017), "Study on fluid-induced vibration power harvesting of square columns under different attack angles", Geofluids, 2017, 1-18. https://doi.org/10.1155/2017/6439401.
  46. Zhang, Y., Ren, W., Li, P., Zhang, X. and Lu, X. (2023), "Flow regimes and characteristics of dense particulate flows with coarse particles in inclined pipe", Powder Technol., 428, 118859. https://doi.org/10.1016/j.powtec.2023.118859.
  47. Zhang, Z., Wang, J., Huang, R., Qiu, R., Chu, X., Ye, S., Wang, Y. and Liu, Q. (2023), "Data-driven turbulence model for unsteady cavitating flow", Phys. Fluids, 35, 015134. https://doi.org/10.1063/5.0134992.
  48. Zhang, Z., Song, X.D., Ye, S.R., Wang, Y.W., Huang, C.F., An, Y.R. and Chen, Y.S. (2019), "Application of deep learning method to Reynolds stress models of channel flow based on reduced-order modeling of DNS data", J Hydrodyn, 31(1), 58-65. https://doi.org/10.1007/s42241-018-0156-9.