Wind and Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

A novel two-layer hybrid model for ultra-short-term wind speed prediction based on SSP and BO-LSTM

  • Weicheng Hu (Zhejiang Jiangnan Project Management Co., Ltd.) ;
  • Baolong Cheng (Zhejiang Jiangnan Project Management Co., Ltd.) ;
  • Qingshan Yang (Chongqing Key Laboratory of Wind Engineering and Wind Energy Utilization, School of Civil Engineering, Chongqing University) ;
  • Zhenqing Liu (School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology) ;
  • Ziting Yuan (School of Civil Engineering and Architecture, East China Jiaotong University) ;
  • Ke Li (Chongqing Key Laboratory of Wind Engineering and Wind Energy Utilization, School of Civil Engineering, Chongqing University) ;
  • Mingjin Zhang (Department of Bridge Engineering, Southwest Jiaotong University)
  • Received : 2022.05.18
  • Accepted : 2022.12.19
  • Published : 2023.05.25
⟨ Previous Next ⟩

Abstract

Grid management is important for energy distribution, system security and market economics, and one of the key issues is accurate and stable prediction of wind speed for optimal operation and management of wind power connected to the grid. In this study, a novel two-layer hybrid method termed SSP-BO-LSTM is proposed for ultra-short-term wind speed prediction, such as four-hour ahead. The first layer is based on the smoothing spline preprocessing (SSP) method to remove non-Gaussian and non-stationary volatilities from the high-resolution wind speed series. Then, the processed wind speed data are predicted four-hour ahead by the long short-term memory (LSTM) model, and a bayesian optimization (BO) algorithm is presented to optimize the hyperparameters of the LSTM model. To evaluate the performance of the proposed SSP-BO-LSTM model, a case study of ultra-short-term wind speed prediction is conducted, including three high-resolution wind speed series from wind turbine measurements. Moreover, six other prediction models are introduced for in-depth comparison, and a comprehensive analysis is performed. The results show that the proposed model can improve the accuracy of four-hour ahead prediction by about 8%-35%, proving to be more effective and stable in providing acceptable results compared to the other six models mentioned in this study.

Keywords

Acknowledgement

The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (Grant No. 52208479), the Jiangxi Provincial Natural Science Foundation Project (Grant No. 20224BAB214070), the China Postdoctoral Science Foundation Project (Grant No. 2022M720577), the Postdoctoral Research Project of Zhejiang Province (Grant No. ZJ2022037), and the Hangzhou Construction Research Project (Grant No. 2022030).

References

  1. Ahmadi, M., and Khashei, M. (2021), "Current status of hybrid structures in wind forecasting", Eng. Appl. Artif. Intel., 99, 104133. https://doi.org/10.1016/j.engappai.2020.104133.
  2. Aly, H.H.H. (2020), "A novel deep learning intelligent clustered hybrid models for wind speed and power forecasting", Energy, 213, 118773. https://doi.org/10.1016/j.energy.2020.118773.
  3. Brown, B.G., Katz, R.W. and Murphy, A.H. (1984), "Time series models to simulate and forecast wind speed and wind power", J. Appl. Meteorol. Climatol., 23(8), 1184-1195. https://doi.org/10.1175/1520-0450(1984)023%3C1184:TSMTSA%3E2.0.CO;2.
  4. Cai, H., Jia, X., Feng, J., Li, W., Hsu, Y.-M. and Lee, J. (2020), "Gaussian process regression for numerical wind speed prediction enhancement", Renew. Energ., 146, 2112-2123. https://doi.org/10.1016/j.renene.2019.08.018.
  5. Cevik, H.H., Cunkas, M. and Polat, K. (2019), "A new multistage short-term wind power forecast model using decomposition and artificial intelligence methods", Phys. A: Stat. Mech. Appl., 534, 122177. https://doi.org/10.1016/j.physa.2019.122177.
  6. Chen, M.-R., Zeng, G.-Q., Lu, K.-D. and Weng, J. (2019), "A two-layer nonlinear combination method for short-term wind speed prediction based on ELM, ENN, and LSTM", IEEE Internet Things J., 6(4), 6997-7010. https://doi.org/10.1109/JIOT.2019.2913176.
  7. Cheng, Z. and Wang, J. (2020), "A new combined model based on multi-objective salp swarm optimization for wind speed forecasting," Appl. Soft Comput., 92, 106294. https://doi.org/10.1016/j.asoc.2020.106294.
  8. Dale, S. (2021), BP Statistical Review of World Energy, BP Pulication, London, UK.
  9. Deng, Y., Wang, B. and Lu, Z. (2020), "A hybrid model based on data preprocessing strategy and error correction system for wind speed forecasting", Energy Conver. Manage., 212, 112779. https://doi.org/10.1016/j.enconman.2020.112779.
  10. Dong, Y., Wang, J., Xiao, L. and Fu, T. (2021), "Short-term wind speed time series forecasting based on a hybrid method with multiple objective optimization for non-convex target", Energy, 215, 119180. https://doi.org/10.1016/j.energy.2020.119180.
  11. Ferreira, M., Santos, A. and Lucio, P. (2019), "Short-term forecast of wind speed through mathematical models", Energy Reports, 5, 1172-1184. https://doi.org/10.1016/j.egyr.2019.05.007.
  12. Guo, Z.H., Wu, J., Lu, H.Y. and Wang, J.Z. (2011), "A case study on a hybrid wind speed forecasting method using BP neural network", Knowl.-Based Syst., 24(7), 1048-1056. https://doi.org/10.1016/j.knosys.2011.04.019.
  13. He, Y. and Tsang, K.F. (2021), "Universities power energy management: A novel hybrid model based on iCEEMDAN and Bayesian optimized LSTM", Energy Reports, 7, 6473-6488. https://doi.org/10.1016/j.egyr.2021.09.115.
  14. 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.
  15. Hong, Y.Y. and Satriani, T.R.A. (2020), "Day-ahead spatiotemporal wind speed forecasting using robust designbased deep learning neural network", Energy, 209, 118441. https://doi.org/10.1016/j.energy.2020.118441.
  16. Hu, J., Wang, J. and Ma, K. (2015), "A hybrid technique for shortterm wind speed prediction", Energy, 81, 563-574. https://doi.org/10.1016/j.energy.2014.12.074.
  17. Hu, Weicheng, Yang, Q., Chen, H.-P., Yuan, Z., Li, C., Shao, S. and Zhang, J. (2021), "New hybrid approach for short-term wind speed predictions based on preprocessing algorithm and optimization theory", Renew. Energ., 179, 2174-2186. https://doi.org/10.1016/j.renene.2021.08.044.
  18. Hu, W., He, Y., Liu, Z., Tan, J., Yang, M. and Chen, J. (2020), "A hybrid wind speed prediction approach based on ensemble empirical mode decomposition and BO-LSTM neural networks for digital twin", ASME Power Conf., 83747, V001T08A009. https://doi.org/10.1115/POWER2020-16500.
  19. Huang, Y., Liu, S. and Yang, L. (2018), "Wind speed forecasting method using EEMD and the combination forecasting method based on GPR and LSTM", Sust., 10(10), 3693. https://doi.org/10.3390/su10103693.
  20. Huang, Z. and Chalabi, Z.S. (1995), "Use of time-series analysis to model and forecast wind speed", J. Wind Eng. Industrial Aerod., 56(2-3), 311-322. https://doi.org/10.1016/0167-6105(94)00093-S.
  21. Hur, S. (2021), "Short-term wind speed prediction using Extended Kalman filter and machine learning", Energy Reports, 7, 1046-1054. https://doi.org/10.1016/j.egyr.2020.12.020.
  22. Jiang, P., Liu, Z., Niu, X. and Zhang, L. (2021), "A combined forecasting system based on statistical method, artificial neural networks, and deep learning methods for short-term wind speed forecasting", Energy, 217, 119361. https://doi.org/10.1016/j.energy.2020.119361.
  23. Kavasseri, R.G. and Seetharaman, K. (2009), "Day-ahead wind speed forecasting using f-ARIMA models", Renew. Energ., 34(5), 1388-1393. https://doi.org/10.1016/j.renene.2008.09.006.
  24. Li, G., Shi, J. and Zhou, J. (2011), "Bayesian adaptive combination of short-term wind speed forecasts from neural network models", Renew. Energ., 36(1), 352-359. https://doi.org/10.1016/j.renene.2010.06.049.
  25. Lin, Z. and Liu, X. (2020), "Wind power forecasting of an offshore wind turbine based on high-frequency SCADA data and deep learning neural network", Energy, 201, 117693. https://doi.org/10.1016/j.energy.2020.117693.
  26. Liu, B., Zhao, S., Yu, X., Zhang, L. and Wang, Q. (2020), "A novel deep learning approach for wind power forecasting based on WD-LSTM model", Energies, 13(18), 4964. https://doi.org/10.3390/en13184964.
  27. Liu, C., Fan, G., Wang, W. and Dai, H. (2009), "A combination forecasting model for wind farm output power", Power Syst. Tech., 33(13), 74-79. https://doi.org/10.13335/j.1000-3673.pst.2009.13.003.
  28. Liu, H., Yu, C., Wu, H., Duan, Z. and Yan, G. (2020), "A new hybrid ensemble deep reinforcement learning model for wind speed short term forecasting", Energy, 202, 117794. https://doi.org/10.1016/j.energy.2020.117794.
  29. Liu, M., Cao, Z., Zhang, J., Wang, L., Huang, C. and Luo, X. (2020), "Short-term wind speed forecasting based on the JayaSVM model, Int. J. Electrical Power Energy Syst., 121, 106056. https://doi.org/10.1016/j.ijepes.2020.106056.
  30. Liu, X., Zhang, H., Kong, X. and Lee, K.Y. (2020), "Wind speed forecasting using deep neural network with feature selection", Neurocomput., 397, 393-403. https://doi.org/10.1016/j.neucom.2019.08.108.
  31. Lu, P., Ye, L., Zhong, W., Qu, Y., Zhai, B., Tang, Y. and Zhao, Y. (2020), "A novel spatio-temporal wind power forecasting framework based on multi-output support vector machine and optimization strategy", J. Cleaner Product., 254, 119993. https://doi.org/10.1016/j.jclepro.2020.119993.
  32. Mahoney, W.P., Parks, K., Wiener, G., Liu, Y., Myers, W.L., Sun, J., Delle Monache, L., Hopson, T., Johnson, D. and Haupt, S.E. (2012), "A wind power forecasting system to optimize grid integration", IEEE Transactions Sust. Energy, 3(4), 670-682. https://doi.org/10.1109/TSTE.2012.2201758.
  33. Memarzadeh, G. and Keynia, F. (2020), "A new short-term wind speed forecasting method based on fine-tuned LSTM neural network and optimal input sets", Energy Conver. Manag., 213, 112824. https://doi.org/10.1016/j.enconman.2020.112824.
  34. Meng, A., Ge, J., Yin, H. and Chen, S. (2016), "Wind speed forecasting based on wavelet packet decomposition and artificial neural networks trained by crisscross optimization algorithm", Energy Conver. Manag., 114, 75-88. https://doi.org/10.1016/j.enconman.2016.02.013.
  35. National Power Dispatch and Communication Center (2015), Functional Specification of Wind Power Forecasting System: Q/GDW 10588-2015, State Grid Corporation of China, Beijing, China.
  36. Qian, W. and Wang, J. (2020), "An improved seasonal GM(1,1) model based on the HP filter for forecasting wind power generation in China", Energy, 209, 118499. https://doi.org/10.1016/j.energy.2020.118499.
  37. Rasmussen, C.E. and Williams, C. (2005), Gaussian Processes for Machine Learning (Volume 1), MIT Press, Cambridge, MA.
  38. Rocha, P.A.C., de Sousa, R.C., de Andrade, C.F. and da Silva, M.E.V. (2012), "Comparison of seven numerical methods for determining Weibull parameters for wind energy generation in the northeast region of Brazil", Appl. Energy, 89(1), 395-400. https://doi.org/10.1016/j.apenergy.2011.08.003.
  39. Moreno, S.R., da Silva, R.G., Mariani, V.C. and dos Santos Coelho, L (2020), "Multi-step wind speed forecasting based on hybrid multi-stage decomposition model and long short-term memory neural network", Energy Conver. Manag., 213, 112869. https://doi.org/10.1016/j.enconman.2020.112869.
  40. Shahriari, B., Swersky, K., Wang, Z., Adams, R.P. and De Freitas, N. (2015), "Taking the human out of the loop: a review of Bayesian optimization", Proceedings of the IEEE, 104(1), 148-175. https://doi.org/10.1109/JPROC.2015.2494218.
  41. Sharifzadeh, M., Sikinioti-Lock, A. and Shah, N. (2019), "Machine-learning methods for integrated renewable power generation: a comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression", Renew. Sust. Energy Rev., 108, 513-538. https://doi.org/10.1016/j.rser.2019.03.040.
  42. Sun, Z., Zhao, M., Dong, Y., Cao, X. and Sun, H. (2021), "Hybrid model with secondary decomposition, randomforest algorithm, clustering analysis and long short memory network principal computing for short-term wind power forecasting on multiple scales", Energy, 221, 119848. https://doi.org/10.1016/j.energy.2021.119848.
  43. Torres, J.L., Garcia, A., De Blas, M. and De Francisco, A. (2005), "Forecast of hourly average wind speed with ARMA models in Navarre (Spain)", Solar Energy, 79(1), 65-77. https://doi.org/10.1016/j.solener.2004.09.013.
  44. Wang, J., Li, Q. and Zeng, B. (2021), "Multi-layer cooperative combined forecasting system for short-term wind speed forecasting", Sust. Energy Tech. Assess., 43, 100946. https://doi.org/10.1016/j.seta.2020.100946.
  45. Wang, W., Wang, Z., Dong, C., Liang, Z., Feng, S. and Wang, B. (2021), "Status and error analysis of short-term forecasting technology of wind power in China", Auto. Electr. Power Syst., 45(1), 17-29. https://doi.org/10.7500/AEPS20200324003.
  46. Wang, Y. and Wu, L. (2016), "On practical challenges of decomposition-based hybrid forecasting algorithms for wind speed and solar irradiation", Energy, 112, 208-220. https://doi.org/10.1016/j.energy.2016.06.075.
  47. Wu, T. and Snaiki, R. (2022), "Applications of machine learning to wind engineering", Front. Built Environ., 8. https://doi.org/10.3389/fbuil.2022.811460.
  48. Xiang, L., Li, J., Hu, A. and Zhang, Y. (2020), "Deterministic and probabilistic multi-step forecasting for short-term wind speed based on secondary decomposition and a deep learning method", Energy Conver. Manag., 220, 113098. https://doi.org/10.1016/j.enconman.2020.113098.
  49. Yan, X., Liu, Y., Xu, Y. and Jia, M. (2020), "Multistep forecasting for diurnal wind speed based on hybrid deep learning model with improved singular spectrum decomposition", Energy Conver. Manag., 225, 113456. https://doi.org/10.1016/j.enconman.2020.113456.
  50. Yang, X., Zhang, Y., Ye, T. and Su, J. (2020), "Prediction of combination probability interval of wind power based on naive Bayes", High Voltage Eng., 46(3), 1099-1108. https://doi.org/10.13336/j.1003-6520.hve.20200331041.
  51. Yildiz, C., Acikgoz, H., Korkmaz, D. and Budak, U. (2021), "An improved residual-based convolutional neural network for very short-term wind power forecasting", Energy Conver. Manag., 228, 113731. https://doi.org/10.1016/j.enconman.2020.113731.
  52. Zhang, J., Meng, H., Gu, B. and Li, P. (2020), "Research on shortterm wind power combined forecasting and its Gaussian cloud uncertainty to support the integration of renewables and EVs", Renew. Energ., 153, 884-899. https://doi.org/10.1016/j.renene.2020.01.062.
  53. Zhang, Y., Chen, B., Pan, G. and Zhao, Y. (2019), "A novel hybrid model based on VMD-WT and PCA-BP-RBF neural network for short-term wind speed forecasting", Energy Conver. Manag., 195, 180-197. https://doi.org/10.1016/j.enconman.2019.05.005.
  54. Zhou, Q., Wang, C. and Zhang, G. (2020), "A combined forecasting system based on modified multi-objective optimization and sub-model selection strategy for short-term wind speed", Appl. Soft Comput., 94, 106463. https://doi.org/10.1016/j.asoc.2020.106463.
  55. Zhu, R., Liao, W. and Wang, Y. (2020), "Short-term prediction for wind power based on temporal convolutional network", Energy Reports, 6, 424-429. https://doi.org/10.1016/j.egyr.2020.11.219.