DOI QR코드

DOI QR Code

Comparison of artificial intelligence models reconstructing missing wind signals in deep-cutting gorges

  • Zhen Wang (School of Civil Engineering, Tianjin University) ;
  • Jinsong Zhu (School of Civil Engineering, Tianjin University) ;
  • Ziyue Lu (Department of Structural Engineering, Norwegian University of Science and Technology) ;
  • Zhitian Zhang (College of Civil Engineering and Architecture, Hainan University)
  • Received : 2023.08.12
  • Accepted : 2024.01.11
  • Published : 2024.01.25

Abstract

Reliable wind signal reconstruction can be beneficial to the operational safety of long-span bridges. Non-Gaussian characteristics of wind signals make the reconstruction process challenging. In this paper, non-Gaussian wind signals are converted into a combined prediction of two kinds of features, actual wind speeds and wind angles of attack. First, two decomposition techniques, empirical mode decomposition (EMD) and variational mode decomposition (VMD), are introduced to decompose wind signals into intrinsic mode functions (IMFs) to reduce the randomness of wind signals. Their principles and applicability are also discussed. Then, four artificial intelligence (AI) algorithms are utilized for wind signal reconstruction by combining the particle swarm optimization (PSO) algorithm with back propagation neural network (BPNN), support vector regression (SVR), long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM), respectively. Measured wind signals from a bridge site in a deep-cutting gorge are taken as experimental subjects. The results showed that the reconstruction error of high-frequency components of EMD is too large. On the contrary, VMD fully extracts the multiscale rules of the signal, reduces the component complexity. The combination of VMD-PSO-Bi-LSTM is demonstrated to be the most effective among all hybrid models.

Keywords

Acknowledgement

Authors would like to express their gratitude to the financial support from the Hainan Provincial Natural Science Foundation of China (Grant Number 520CXTD433); They are also indebted to the National Natural Science Foundation of China (Grant Number 51938012, 52268073 and 52068020).

References

  1. Antonanzas-Torres, F., Urraca, R. and Antonanzas, J., Fernandez-Ceniceros, J. and Martinez-de-Pison, F. (2015), "Generation of daily global solar irradiation with support vector machines for regression", Energ. Convers. Manage., 96, 277-286. https://doi.org/10.1016/j.enconman.2015.02.086.
  2. Barbounis, T., Theocharis, J., Alexiadis, M. and Dokopoulos, P. (2006), "Long-term wind speed and power forecasting using local recurrent neural network models", IEEE T. Energy Conver., 21(1), 273-284. https://doi.org/10.1109/TEC.2005.847954.
  3. Cadenas, E. and Rivera, R. (2010), "Wind speed forecasting in three different regions of Mexico, using a hybrid ARIMA-ANN model", Renew. Energ., 35(12), 2732-2738. https://doi.org/10.1016/j.renene.2010.04.022.
  4. Dragomiretskiy, K. and Zosso, D. (2014), "Variational Mode Decomposition", IEEE T. Signal Proces., 62(3), 531-544. https://doi.org/10.1109/TSP.2013.2288675
  5. Fu, W., Wang, K., Tan, J. and Zhang, K. (2020), "A composite framework coupling multiple feature selection, compound prediction models and novel hybrid swarm optimizer-based synchronization optimization strategy for multi-step ahead short-term wind speed forecasting", Energ. Convers. Manage., 205, 112461. https://doi.org/10.1016/j.enconman.2019.112461.
  6. Fuentealba, P., Illanes, A. and Ortmeier, F. (2019), "Cardiotocographic signal feature extraction through CEEMDAN and time-varying autoregressive spectral-based analysis for fetal welfare assessment", IEEE Access, 7(1), 159754-159772. https://doi.org/10.1109/ACCESS.2019.2950798.
  7. Georgilakis, P. (2008), "Technical challenges associated with the integration of wind power into power systems", Renew. Sust. Energ. Rev., 12(3), 852-863. https://doi.org/10.1016/j.rser.2006.10.007.
  8. He J., Yu C., Li Y. and Xiang H. (2020), "Ultra-short term wind prediction with wavelet transform, deep belief network and ensemble learning", Energ. Convers. Manage., 205, 112418. https://doi.org/10.1016/j.enconman.2019.112418.
  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. Hong, Y. and Rioflorido, C. (2019), "A hybrid deep learning-based neural network for 24-h ahead wind power forecasting", Appl. Energ., 250, 530-539. https://doi.org/10.1016/j.apenergy.2019.05.044.
  11. Hu, J., Wang, J. and Zeng, G. (2013), "A hybrid forecasting approach applied to wind speed time series", Renew. Energ., 60, 185-194. https://doi.org/10.1016/j.renene.2013.05.012.
  12. Hu W.C., Cheng B.L., Yang Q.S., Liu Z.Q., Yuan Z.T., Li, K. and Zhang, M.J. (2023), "A novel two-layer hybrid model for ultra-short-term wind speed prediction based on SSP and BOLSTM", Wind Struct., 36(5), 293-305. https://doi.org/10.12989/was.2023.36.5.293.
  13. Huang, N., Shen, Z., Long, S., Wu, M., Shih, H., Zheng, Q., Yen, N., Tung, C. and Liu, H. (1998), "The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis", Proc. R. Soc. A-Math. Phy. Eng. Sci., 454(1971), 903-995. https://doi.org/10.1098/rspa.1998.0193
  14. Jiang P. and Li C. (2018), "Research and application of an innovative combined model based on a modified optimization algorithm for wind speed forecasting", Measurement, 124, 395-412. https://doi.org/10.1016/j.measurement.2018.04.014.
  15. Karim, F., Majumdar, S. and Darabi, H. (2019), "Insights into LSTM Fully Convolutional Networks for Time Series Classification", IEEE Access, 7, 67718-67725, https://doi.org/10.1109/ACCESS.2019.2916828.
  16. Kennedy, J. and Eberhart, R. (1995), "Particle Swarm Optimization", Icnn95-international Conference on Neural Networks. IEEE.
  17. Khosravi, A., Machado, L. and Nunes, R. (2018), "Time-series prediction of wind speed using machine learning algorithms, A case study Osorio wind farm", Brazil. Appl. Energ., 224, 550-566. https://doi.org/10.1016/j.apenergy.2018.05.043.
  18. Kiplangat, D., Asokan, K. and Kumar, K. (2016), "Improved week-ahead predictions of wind speed using simple linear models with wavelet decomposition", Renew. Energ., 93, 38-44, https://doi.org/10.1016/j.renene.2016.02.054.
  19. Li, L., Zhao, X., Tseng, M., L. and Tan, R. (2020(a)), "Short-term wind power forecasting based on support vector machine with improved dragonfly algorithm", J. Clean. Prod., 242, 118447. https://doi.org/10.1016/j.jclepro.2019.118447.
  20. Li, J.L., Song, Z.H., Wang, X.F., Wang, Y.R. and Jia, Y.Y. (2022), "A novel offshore wind farm typhoon wind speed prediction model based on PSO-Bi-LSTM improved by VMD", Energy, 251, 123848. https://doi.org/10.1016/j.energy.2022.123848.
  21. Li, S., Liu, X. and Lin, A., (2020(b)), "Fractional frequency hybrid model based on EEMD for financial time series forecasting", Commun. Nonlinear Sci. Numer. Simul., 89, 105281. https://doi.org/10.1016/j.cnsns.2020.105281.
  22. Li, W., Jia, X., Li, X., Wang, Y. and Lee, J. (2021), "A Markov model for short term wind speed prediction by integrating the wind acceleration information", Renew. Energ., 164, 242-53. https://doi.org/10.1016/j.renene.2020.09.031.
  23. Li, Y., Yin, J.T., Chen, F.B. and Li, Q.S. (2023), "Machine learning-based prediction of wind forces on CAARC standard tall buildings", Wind Struct., 36(6) 355-366. https://doi.org/10.12989/was.2023.36.6.355.
  24. Liao, H., Jing, H., Ma, C., Tao, Q. and Li, Z., (2020), "Field measurement study on turbulence field by wind tower and Windcube Lidar in mountain valley", J. Wind Eng. Ind. Aerod., 197, 104090. https://doi.org/10.1016/j.jweia.2019.104090.
  25. Lim, J., Kim, S., Kim, H. and Kim, Y. (2022), "Long short-term memory (LSTM)-based wind speed prediction during a typhoon for bridge traffic control", J. Wind Eng. Ind. Aerod., 220, 104778. https://doi.org/10.1016/j.jweia.2021.104788
  26. Liu, H., Duan, Z., Wu, H.P., Li, Y.F. and Dong, S.Y. (2019(a)), "Wind speed forecasting models based on data decomposition, feature selection and group method of data handling network", Measurement, 148, 106971. https://doi.org/10.1016/j.measurement.2019.106971.
  27. Liu, H., Tian, H., Pan, D. and Li, Y. (2013), "Forecasting models for wind speed using wavelet, wavelet packet, time series and Artificial Neural Networks", Appl. Energy., 107, 191-208. https://doi.org/10.1016/j.apenergy.2013.02.002.
  28. Liu, H., Yang, R., Wang, T. and Zhang, L. (2021), "A hybrid neural network model for short-term wind speed forecasting based on decomposition, multi-learner ensemble, and adaptive multiple error corrections", Renew. Energ., 165, 573-594. https://doi.org/10.1016/j.renene.2020.11.002.
  29. Liu, H., Mi, X., Li, Y., Duan, Z. and Xu, Y. (2019(b)), "Smart wind speed deep learning based multi-step forecasting model using singular spectrum analysis, convolutional gated recurrent unit network and support vector regression", Renew. Energ., 143, 842-854. https://doi.org/10.1016/j.renene.2019.05.039.
  30. Liu, M., Cao, Z., Zhang, J., Wang, L., Huang, C. and Luo X. (2020), "Short-term wind speed forecasting based on the Jaya-SVM model", Int. J. Electr. Power Energy Syst., 121, 106056. https://doi.org/10.1016/j.ijepes.2020.106056.
  31. Noorollahi, Y., Jokar, M. and Kalhor, A. (2016), "Using artificial neural networks for temporal and spatial wind speed forecasting in Iran", Energ. Convers. Manage., 115, 17-25. https://doi.org/10.1016/j.enconman.2016.02.041.
  32. Pei, S., Qian, H., Zhang, Z., Yao, L., Wang, Y., Wang, C., Liu, Y., Jiang, Z., Zhou, J. and Yi, T. (2019), "Wind speed prediction method based on Empirical Wavelet Transform and New Cell Update Long Short-Term Memory network", Energ. Convers. Manage., 196, 779-792. https://doi.org/10.1016/j.enconman.2019.06.041.
  33. Qian, Z., Pei, Y., Zareipour, H. and Chen, N.Y. (2019), "A review and discussion of decomposition-based hybrid models for wind energy forecasting applications", Appl. Energ., 235, 939-953. https://doi.org/10.1016/j.apenergy.2018.10.080.
  34. Ren, Y., Suganthan, P. and Srikanth, N. (2016), "A novel empirical mode decomposition with support vector regression for wind speed forecasting", IEEE T. Neur. Net. Lear Syst., 27(8), 1793-1798. https://doi.org/10.1109/TNNLS.2014.2351391.
  35. Rumelhart, D. and Mcclelland, J. (1986), Parallel Distributed Processing, MIT Press, Cambridge, USA.
  36. Smola, A. and Lkopf, B. (2004), "A tutorial on support vector regression", Stat. Comput., 14(3), 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88.
  37. Tascikaraoglu, A. and Uzunoglu, M. (2014), "A review of combined approaches for prediction of short-term wind speed and power", Renew. Sust. Energ. Rev., 34, 243-254. https://doi.org/10.1016/j.rser.2014.03.033.
  38. Vapnik, V. (1995), The Nature of Statistical Learning Theory, Springer, New York, USA.
  39. Wang, S., Zhang, N., Wu, L. and Wang, Y. (2016), "Wind speed forecasting based on the hybrid ensemble empirical mode decomposition and GA-BP neural network method", Renew. Energ., 94, 629-636. https://doi.org/10.1016/j.renene.2016.03.103.
  40. Yu, C., Li, Y., Xiang, H. and Zhang, M. (2018), "Data mining-assisted short-term wind speed forecasting by wavelet packet decomposition and Elman neural network", J. Wind Eng. Ind. Aerod., 178, 136-143. https://doi.org/10.1016/j.jweia.2018.01.020.
  41. Yu, C., Li, Y., Zhao, L., Qian, C. and Xun, Y. (2023), "A novel time-frequency recurrent network and its advanced version for short-term wind speed predictions", Energy, 262, 125556. https://doi.org/10.1016/j.energy.2022.125556.
  42. Zhang, C., Li, R., Shi, H. and Li, F. (2020(a)), "Deep learning for day-ahead electricity price forecasting", IET Smart Grid, 3(4), 462-469. https://doi.org/10.1049/iet-stg.2019.0258.
  43. Zhang, C., Wei, H., Zhao, J., Liu, T., Zhu, T. and Zhang, K. (2016(a)), "Short-term wind speed forecasting using empirical mode decomposition and feature selection", Renew. Energ., 96, 727-737. https://doi.org/10.1016/j.renene.2016.05.023.
  44. Zhang, C., Wei, H., Zhao, X., Liu, T. and Zhang K. (2016(b)), "A Gaussian process regression based hybrid approach for short-term wind speed prediction", Energ. Convers. Manage., 126, 1084-1092. https://doi.org/10.1016/j.enconman.2016.08.086.
  45. Zhang, Y., Chen, B., Pan, G. and Zhao, Y. (2019(a)), "A novel hybrid model based on VMD-WT and PCA-BP-RBF neural network for short-term wind speed forecasting", Energ. Convers. Manage., 195, 180-197, https://doi.org/10.1016/j.enconman.2019.05.005.
  46. Zhang, Y., Pan, G., Chen, B., Han, J., Zhao, Y. and Zhang, C. (2020(b)), "Short-term wind speed prediction model based on GA-ANN improved by VMD", Renew. Energ., 156, 1373-1388. https://doi.org/10.1016/j.renene.2019.12.047.
  47. Zhang, Y., Zhao, Y. and Gao, S. (2019(b)), "A novel hybrid model for wind speed prediction based on VMD and neural network considering atmospheric uncertainties", IEEE Access, 7, 60322-60332. https://doi.org/10.1109/ACCESS.2019.2915582.
  48. Zhang, Z., Ye, L., Qin, H., Liu, Y., Wang, C., Yu, X., Yin, X. and Li, J. (2019(c)), "Wind speed prediction method using shared weight long short-term memory network and gaussian process regression", Appl. Energ., 247, 270-284. https://doi.org/10.1016/j.apenergy.2019.04.047.
  49. Zhao, X., Liu, J., Yu, D. and Chang, J. (2018), "One-day-ahead probabilistic wind speed forecast based on optimized numerical weather prediction data", Energ. Convers. Manage., 164, 560-569. https://doi.org/10.1016/j.enconman.2018.03.030.
  50. Zhou, J., Shi, J. and Li, G. (2011), "Fine tuning support vector machines for short-term wind speed forecasting", Energ. Convers. Manage., 52(4), 1990-1998, https://doi.org/10.1016/j.enconman.2010.11.007.
  51. Zhou, Y. and Huang, M. (2016), "Lithium-ion batteries remaining useful life prediction based on a mixture of empirical mode decomposition and ARIMA model", Microelectron. Reliab., 65, 265-273. https://doi.org/10.1016/j.microrel.2016.07.151.