Efficient and accurate analysis of wind resources is essential for offshore wind power assessment and energy system dispatch. Different from single-variable prediction, this work proposes a multi-stage multivariable offshore wind speed assessment system. Firstly, the main variable is decomposed into three terms using seasonal-trend decomposition with loess (STL). Six highly correlated variables are then selected from 13 wind speed factors using a combination of three methods. Prediction model is built using Bidirectional Long Short-Term Memory networks (BiLSTM) based on Multi-Head Self-Attention (MSA) mechanisms, wherein Convolutional Neural Networks (CNNs) and Gated Recurrent Units (GRUs) are added. Sub-term prediction biases are corrected using Artificial Neural Networks (ANNs) optimized by the Dung Beetle Optimizer (DBO). Wind speed for four specific months representing the four seasons of an offshore wind farm is computed. Comparative experiments show that the proposed method achieves excellent performance with high prediction accuracy.
AasimSinghSNMohapatraA (2019) Repeated wavelet transform based ARIMA model for very short-term wind speed forecasting. Renewable Energy136: 758–768. DOI: 10.1016/j.renene.2019.01.031.
2.
AbbasimehrHPakiRBahriniA (2023) A novel XGBoost-based featurization approach to forecast renewable energy consumption with deep learning models. Sustainable Computing: Informatics and Systems38: 100863. DOI: 10.1016/j.suscom.2023.100863.
3.
AlexiadisMCDokopoulosPSSahsamanoglouHS (1999) Wind speed and power forecasting based on spatial correlation models. IEEE Transactions on Energy Conversion14(3): 836–842. DOI: 10.1109/60.790962.
4.
AliYAlyHH (2024) Short term wind speed forecasting using artificial and wavelet neural networks with and without wavelet filtered data based on feature selections technique. Engineering Applications of Artificial Intelligence133: 108201. DOI: 10.1016/j.engappai.2024.108201.
5.
AssarehEBehrangMAGhalambazM, et al. (2012) An analysis of wind speed prediction using artificial neural networks: a case study in Manjil, Iran. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects34(7): 636–644. DOI: 10.1080/15567036.2011.551915.
6.
BergmeirCHyndmanRJBenítezJM (2016) Bagging exponential smoothing methods using STL decomposition and Box–Cox transformation. International Journal of Forecasting32(2): 303–312. DOI: 10.1016/j.ijforecast.2015.07.002.
7.
BouabdallaouiDHaidiTElmariamiF, et al. (2023) Application of four machine-learning methods to predict short-horizon wind energy. Global Energy Interconnection6(6): 726–737. DOI: 10.1016/j.gloei.2023.11.006.
8.
ButuracheAStancuS (2021) Wind energy prediction using machine learning. Low Carbon Economy12: 1–21. DOI: 10.4236/lce.2021.121001.
9.
CadenasERiveraW (2010) Wind speed forecasting in three different regions of Mexico, using a hybrid ARIMA–ANN model. Renewable Energy35(12): 2732–2738. DOI: 10.1016/j.renene.2010.04.022.
10.
CarrilloCCidrásJDíaz-DoradoE, et al. (2014) An approach to determine the Weibull parameters for wind energy analysis: the case of Galicia (Spain). Energies7(4): 2676–2700. DOI: 10.3390/en7042676.
11.
ChenHChangX (2021) Photovoltaic power prediction of LSTM model based on Pearson feature selection. Energy Reports7: 1047–1054. DOI: 10.1016/j.egyr.2021.09.167.
12.
CunhaJLRNPereiraCMNA (2024) A hybrid model based on STL with simple exponential smoothing and ARMA for wind forecast in a Brazilian nuclear power plant site. Nuclear Engineering and Design421: 113026. DOI: 10.1016/j.nucengdes.2024.113026.
13.
DragomiretskiyKZossoD (2014) Variational mode decomposition. IEEE Transactions on Signal Processing62(3): 531–544. DOI: 10.1109/TSP.2013.2288675.
14.
EsenHInalliMSengurA, et al. (2008) Forecasting of a ground-coupled heat pump performance using neural networks with statistical data weighting pre-processing. International Journal of Thermal Sciences47(4): 431–441. DOI: 10.1016/j.ijthermalsci.2007.03.004.
15.
FangYWuYWuF, et al. (2023) Short-term wind speed forecasting bias correction in the Hangzhou area of China based on a machine learning model. Atmospheric and Oceanic Science Letters16(4): 100339. DOI: 10.1016/j.aosl.2023.100339.
16.
HongYYChangHLChiuCS (2010) Hour-ahead wind power and speed forecasting using Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm and neural network with fuzzy inputs. Energy35(9): 3870–3876. DOI: 10.1016/j.energy.2010.05.041.
17.
JiangJZhangXYuanZ (2024) Feature selection for classification with Spearman’s rank correlation coefficient-based self-information in divergence-based fuzzy rough sets. Expert Systems with Applications249: 123633. DOI: 10.1016/j.eswa.2024.123633.
18.
JosephLPDeoRCCasillas-PérezD, et al. (2024) Short-term wind speed forecasting using an optimized three-phase convolutional neural network fused with bidirectional long short-term memory network model. Applied Energy359: 122624. DOI: 10.1016/j.apenergy.2024.122624.
LiGShiJ (2010) On comparing three artificial neural networks for wind speed forecasting. Applied Energy87(7): 2313–2320. DOI: 10.1016/j.apenergy.2009.12.013.
21.
LiuDNiuDWangH, et al. (2014) Short-term wind speed forecasting using wavelet transform and support vector machines optimized by genetic algorithm. Renewable Energy62: 592–597. DOI: 10.1016/j.renene.2013.08.011.
22.
LiuHWuHLiY (2020) Multi-step wind speed forecasting model based on wavelet matching analysis and hybrid optimization framework. Sustainable Energy Technologies and Assessments40: 100745. DOI: 10.1016/j.seta.2020.100745.
23.
LuengoJGarcíaSHerreraF (2012) On the choice of the best imputation methods for missing values considering three groups of classification methods. Knowledge and Information Systems32: 77–108. DOI: 10.1007/s10115-011-0424-2.
24.
LvS-XWangL (2023) Multivariate wind speed forecasting based on multi-objective feature selection approach and hybrid deep learning model. Energy263: 126100. DOI: 10.1016/j.energy.2022.126100.
25.
MiraftabzadehSMLongoM (2023) High-resolution PV power prediction model based on the deep learning and attention mechanism. Sustainable Energy, Grids and Networks34: 101025. DOI: 10.1016/j.segan.2023.101025.
26.
NiuZZhongGYuH (2021) A review on the attention mechanism of deep learning. Neurocomputing452: 48–62. DOI: 10.1016/j.neucom.2021.03.091.
27.
NiuYWangJZhangZ, et al. (2024) A combined interval prediction system based on fuzzy strategy and neural network for wind speed. Applied Soft Computing155: 111408. DOI: 10.1016/j.asoc.2024.111408.
28.
OktaviaSSyafrianiDLDwiridalL, et al. (2023) Analysis of surface wind speed at Minangkabau international airport for the period 2011–2020 using the Windrose method. Journal of Physics: Conference Series2582(1): 012006. DOI: 10.1088/1742-6596/2582/1/012006.
29.
RanjanKGPrustyBRJenaD (2021) Review of preprocessing methods for univariate volatile time-series in power system applications. Electric Power Systems Research191: 106885. DOI: 10.1016/j.epsr.2020.106885.
30.
SanthoshMVenkaiahCKumarDV (2019) Short-term wind speed forecasting approach using ensemble empirical mode decomposition and deep Boltzmann machine. Sustainable Energy, Grids and Networks19: 100242. DOI: 10.1016/j.segan.2019.100242.
31.
SweeneyCLynchPNolanP (2013) Reducing errors of wind speed forecasts by an optimal combination of post-processing methods. Meteorological Applications20: 32–40. DOI: 10.1002/met.294.
32.
XuLOuYCaiJ, et al. (2023) Offshore wind speed assessment with statistical and attention-based neural network methods based on STL decomposition. Renewable Energy216: 119097. DOI: 10.1016/j.renene.2023.119097.
33.
XueJShenB (2023) Dung beetle optimizer: a new meta-heuristic algorithm for global optimization. The Journal of Supercomputing79: 7305–7336. DOI: 10.1007/s11227-022-04959-6.
34.
YangSDengZLiX, et al. (2021) A novel hybrid model based on STL decomposition and one-dimensional convolutional neural networks with positional encoding for significant wave height forecast. Renewable Energy173: 531–543. DOI: 10.1016/j.renene.2021.04.010.
35.
YinHJinDGuYH, et al. (2020) STL-ATTLSTM: vegetable price forecasting using STL and attention mechanism-based LSTM. Agriculture10(12): 612. DOI: 10.3390/agriculture10120612.
36.
YuzgecUDokurEBalciM (2024) A novel hybrid model based on empirical mode decomposition and echo state network for wind power forecasting. Energy300: 131546. DOI: 10.1016/j.energy.2024.131546.
37.
ZhangJCuiMHodgeB, et al. (2017) Ramp forecasting performance from improved short-term wind power forecasting over multiple spatial and temporal scales. Energy122: 528–541. DOI: 10.1016/j.energy.2017.01.104.
38.
ZhangYHanJPanG, et al. (2021) A multi-stage predicting methodology based on data decomposition and error correction for ultra-short-term wind energy prediction. Journal of Cleaner Production292: 125981. DOI: 10.1016/j.jclepro.2021.125981.
