To address the limitations of existing short-term wind speed prediction methods in terms of accuracy, this study proposes a hybrid prediction model that integrates the variational mode decomposition (VMD), the Newton-Raphson-based optimizer (NRBO), whale optimization algorithm (WOA), and the extreme learning machine (ELM). The penalty factor α and mode number K of VMD are dynamically optimized by NRBO, yielding intrinsic mode functions (IMFs) with well-separated center frequencies and narrow bandwidths, effectively suppressing mode mixing and improving the learning efficiency and prediction accuracy of each sub-sequence. The decomposition produces four high-frequency components (IMF1~IMF4, accounting for 4.95% of total energy) and one low-frequency trend component (IMF5, accounting for 95.05% of total energy), where IMF5ʼs prediction accuracy primarily determines the upper bound of the overall model error through error propagation. The weights and thresholds of ELM for each component are subsequently optimized by WOA to enhance individual component prediction and final signal reconstruction. Validated using field-measured wind speed data at a bridge site, the results demonstrate that, relative to the baseline model WOA-ELM, the introduction of VMD decomposition reduces EMAE by 60.09%, and the subsequent incorporation of NRBO further reduces EMAE by 41.67%. Compared with four conventional single models, EMAE reductions of 80.11%, 80.53%, 63.79%, and 80.37% are achieved, respectively. Furthermore, WOA optimization of ELM weights and thresholds improves model stability and generalization capability, confirming the effectiveness of the proposed parameter dynamic optimization strategy with demonstrated potential for engineering applications.
为直观对比模型预测效果,验证模型有效性,本文选取平均绝对误差(mean absolute error,MAE)EMAE、均方根误差(root mean square error,RMSE)ERMSE和决定系数(coefficient of determination,R2)三项指标评估预测模型性能. 假设真实值为,预测值为,共有n个数据点,真实值的均值为,各误差的表 达式如下:
WANGN, LUOR B, LIAOJ,et al .Short-term wind speed prediction based on wavelet packet decomposition and BP neural network[J].Control and Information Technology,2019(4):44-49.(in Chinese)
GUOS N, MAX, YANGF,et al .A short-term wind speed prediction method based on optimal Copula correlation analysis[J]. Power System and Clean Energy,2024,40(2):63-72.(in Chinese)
LAIMAS H, LIW J, FENGH,et al .Research progress of machine learning in bridge wind engineering[J]. China Journal of Highway and Transport,2023,36(8): 1-13.(in Chinese)
ZUOF Y, ZHANGY Q, ZHAOQ,et al. Two-stage stochastic optimization for operation scheduling and capacity allocation of integrated energy production unit considering supply and demand uncertainty[J]. Proceedings of the CSEE, 2022, 42(22): 8205-8214.(in Chinese)
[14]
KUSIAKA, ZHANGZ J, VERMAA. Prediction,operations, and condition monitoring in wind energy[J]. Energy, 2013, 60: 1-12.
[15]
FUENTESM, CHENL, DAVISJ M,et al .Modeling and predicting complex space-time structures and patterns of coastal wind fields[J].Environmetrics, 2005, 16(5): 449-464.
[16]
WANGX L, WUJ .Active power smoothing of wind generator based on wind speed prediction and inverse-system[J]. Advanced Materials Research,2011,383/384/385/386/387/388/389/390:1363-1368.
[17]
LIS. Wind power prediction using recurrent multilayer perceptron neural networks[C]//IEEE Power Engineering Society General Meeting. 2003: 2325-2330.
[18]
LIUH, MIX W, LIY F .Wind speed forecasting method based on deep learning strategy using empirical wavelet transform,long short term memory neural network and Elman neural network[J].Energy Conversion and Management,2018,156:498-514.
[19]
HUQ H, ZHANGS G, XIEZ X,et al. Noise model based ν- support vector regression with its application to short-term wind speed forecasting[J].Neural Networks,2014,57:1-11.
[20]
ELLISN, DAVYR, TROCCOLIA. Predicting wind power variability events using different statistical methods driven by regional atmospheric model output[J]. Wind Energy, 2015, 18(9): 1611-1628.
[21]
YUC J, LIY L, BAOY L,et al .A novel framework for wind speed prediction based on recurrent neural networks and support vector machine[J].Energy Conversion and Management,2018,178:137-145.
[22]
HUANGG B, ZHUQ Y, SIEWC K .Extreme learning machine:theory and applications[J]. Neurocomputing,2006,70(1/2/3):489-501.
[23]
OKUMUSI, DINLERA. Current status of wind energy forecasting and a hybrid method for hourly predictions[J]. Energy Conversion and Management,2016,123:362-371.
[24]
HERVÁS-MARTÍNEZC, GUTIÉRREZP A, FERNÁNDEZJ C,et al .Hyperbolic tangent basis function neural networks training by hybrid evolutionary programming for accurate short-term wind speed prediction[C]//2009 Ninth International Conference on Intelligent Systems Design and Applications. November 30-December 2,2009.Pisa,Italy.IEEE,2009: 193-198.
[25]
WANGJ Z, QINS S, ZHOUQ P,et al. Medium-term wind speeds forecasting utilizing hybrid models for three different sites in Xinjiang,China[J].Renewable Energy, 2015, 76: 91-101.
[26]
DONGY, NIUJ, LIUQ,et al .A hybrid prediction model for wind speed using support vector machine and genetic programming in conjunction with error compensation[J]. Stochastic Environmental Research and Risk Assessment,2021,35(12): 2411-2424.
[27]
TAOJ, WEIJ N, ZHOUH J,et al .Enhanced prediction model of short-term sea surface wind speed:a multiscale feature extraction and selection approach coupled with deep learning technique[J].Journal of King Saud University-Computer and Information Sciences, 2024, 36(8): 102192.
GAOS Y, LIF S .Short-term wind speed forecasting based on CEEMD-SE-PSR-BP method[J].Acta Energiae Solaris Sinica,2025,46(4):415-422.(in Chinese)
[30]
MIRJALILIS, LEWISA .The whale optimization algorithm[J].Advances in Engineering Software,2016,95:51-67.
[31]
SHENW G, YINGZ Q, ZHAOY M,et al .Significant wave height prediction in monsoon regions based on the VMD-CNN-BiLSTM model[J]. Frontiers in Marine Science,2024,11:1503552.
[32]
SOWMYAR, PREMKUMARM, JANGIRP .Newton-Raphson-based optimizer:a new population-based metaheuristic algorithm for continuous optimization problems[J]. Engineering Applications of Artificial Intelligence,2024,128:107532.
[33]
LIUY K, GUY, LONGY W,et al .Research on physically constrained VMD-CNN-BiLSTM wind power prediction[J]. Sustainability,2025,17(3): 1058.