海浪有效波高具有显著随机性,难以满足高精度预测的实际需求。为此,提出一种基于傅里叶变换与多头轴向注意力机制（FFT-MaxVIT）融合模型对海浪有效波高预测的方法。首先,采用快速傅里叶变换（FFT）突出波浪的主要频率分量,并抑制噪声干扰;其次,在小数据样本的前提下,既通过卷积层提取局部特征,又借助块注意力和网格注意力在降低复杂度的同时高效提取全局特征;最后,使用剪枝技术和贝叶斯优化算法对模型调参,高效地寻找最优超参数组合。试验中,先进行了FFT前后对比性分析,又将该文提出的模型与长短时记忆网络（LSTM）、残差网络（ResNet）和Vision Transformer（ViT）模型预测结果进行对比性分析。实验结果表明,该文提出的模型有效提升了有效波高的预测精度。
Abstract
The significant randomness of the effective wave height of ocean waves makes it difficult to meet the actual demand for high-precision prediction. A novel method for predicting effective wave height based on a fusion model combining the fast Fourier transform （FFT） with a multi-head axial attention mechanism （FFT-MaxVIT） is proposed in this paper. Firstly, the fast Fourier transform （FFT） is employed to accentuate the dominant frequency components of the waves while suppressing noise interference. secondly, specifically addressing scenarios involving small data samples, the model utilizes convolutional layers to extract local features; concurrently, it leverages both block attention and grid attention mechanisms to efficiently extract global features while effectively reducing computational complexity. Finally, pruning techniques and Bayesian optimization algorithms are applied to fine-tune the model parameters, thereby efficiently identifying the optimal combination of hyperparameters. The experimental evaluation involved two stages： first, a comparative analysis was conducted to assess the impact of applying the FFT; second, the predictive performance of the proposed model was benchmarked against that of several established models, including the long short-term memory network （LSTM）, the residual network（ResNet）, and the Vision Transformer （ViT）. The experimental results demonstrate that the proposed model effectively enhances the prediction accuracy of effective wave height.
关键词
有效波高 /
FFT /
贝叶斯优化 /
海浪 /
深度学习 /
预测
Key words
significant wave height /
FFT /
Bayesian optimization /
waves /
deep learning /
prediction
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 夏海南, 王项南, 李强, 等. 波浪能发电装置现场测试中波浪参数比测分析[J]. 太阳能学报, 2022, 43(6): 251-255.
Xia H N, Wang X N, Li Q, et al.Comparison and analysis of wave parameters in field test of wave energy converters[J]. Acta energiae solaris sinica, 2022, 43(6): 251-255.
[2] 王春晓, 于华明, 李松霖, 等. 基于海浪再分析数据的波浪能资源分析[J]. 太阳能学报, 2022, 43(9): 430-436.
Wang C X, Yu H M, Li S L, et al.Wave energy resource valuation based on sea wave reanalysis data[J]. Acta energiae solaris sinica, 2022, 43(9): 430-436.
[3] Hasselmann K F, Barnett T, Bouws E, et al.Measurements of wind-wave growth and swell decay during the Joint North Sea Wave Project (JONSWAP)[J]. Ergaenzungsheft zur Deutschen Hydrographischen Zeitschrift. 1973
[4] Booij N, Ris R C, Holthuijsen L H.A third-generation wave model for coastal regions: 1. Model description and validation[J]. Journal of geophysical research: oceans, 1999, 104(C4): 7649-7666.
[5] TOlman H L, Balasubramaniyan B, Burroughs L D, et al.Development and implementation of wind-generated ocean surface wave Modelsat NCEP[J]. Weather and forecasting, 2002, 17(2): 311-333.
[6] DEo M C, Sridhar Naidu C.Real time wave forecasting using neural networks[J]. Ocean engineering, 1998, 26(3): 191-203.
[7] Makarynskyy O.Improving wave predictions with artificial neural networks[J]. Ocean engineering, 2004, 31(5/6): 709-724.
[8] 张振全, 李醒飞, 杨少波. 基于AR-SVR模型的有效波高短期预测[J]. 太阳能学报, 2021, 42(7): 15-20.
Zhang Z Q, Li X F, Yang S B.Short-term prediction of significant wave height based on AR-SVR model[J]. Acta energiae solaris sinica, 2021, 42(7): 15-20.
[9] Fan S T, Xiao N H, Dong S.A novel model to predict significant wave height based on long short-term memory network[J]. Ocean engineering, 2020, 205: 107298.
[10] Zhou S Y, Xie W H, Lu Y X, et al.ConvLSTM-based wave forecasts in the south and East China Seas[J]. Frontiers in marine science, 2021, 8: 680079.
[11] Pokhrel P, Ioup E, Simeonov J, et al.A transformer-based regression scheme for forecasting significant wave heights in oceans[J]. IEEE journal of oceanic engineering, 2022, 47(4): 1010-1023.
[12] Jing Y, Zhang L, Hao W, et al.Numerical study of a CNN-based model for regional wave prediction[J]. Ocean engineering, 2022, 255: 111400.
[13] Liu Y L, Huang L M, Ma X W, et al.A fast, high-precision deep learning model for regional wave prediction[J]. Ocean engineering, 2023, 288: 115949.
[14] Lyu Y Q, Jiang J, Zhang K, et al.Factorizing and reconstituting large-kernel MBConv for lightweight face recognition[C]//2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Seoul, Korea, 2020: 2689-2697.
[15] Tu Z Z, Talebi H, Zhang H, et al.MaxViT: multi-axis vision transformer[C]//Computer Vision - ECCV 2022. Cham: Springer, 2022: 459-479.
基金
国家自然科学基金（U24A20137; 52405610）
PDF(961 KB)
