基于双向长短期记忆神经网络（Bi-LSTM）建立针对于10 MW漂浮式海上风电平台在波浪作用下的平台运动响应预测模型,通过仿真计算得到大量波浪时间序列信息以及运动响应数据,针对这些数据进行参数敏感性分析,训练后优化参数以确定最优的Bi-LSTM神经网络结构。结果表明,通过考虑不同波高和谱峰频率的波浪条件,验证了Bi-LSTM神经网络方法的可行性。所建立的Bi-LSTM模型对预测输入数据1/3时长的漂浮式海上风电平台在波浪作用下的运动具有较高的准确率,纵荡、垂荡和纵摇的预报精度高达95%,因此所提方法具有较强的平台运动预测能力。
Abstract
This study has developed a method for the motion prediction of a 10 MW floating wind platform under the action of waves, based on the Bi-directional long-short-term memory（Bi-LSTM） neural network. By simulating a 10 MW floating offshore wind power platform, wave and motion time series are obtained for a parameter sensitivity analysis. The simulation data are used to train the Bi-LSTM neural network framework and the parameters are then optimized. The results show that the developed Bi-LSTM model is highly effective in predicting the motion of the floating offshore wind platform under wave action in the next 1/3 time-length of the input data considering different wave heights and spectral peak frequencies. The prediction accuracy of the wave-induced heave and surge is as high as 95%. Therefore, the method proposed in this study has a strong ability to predict platform motion and is of great importance for the development of offshore wind energy.
关键词
漂浮式风电平台 /
深度学习 /
Bi-LSTM /
运动预测 /
申请网络 /
波浪载荷
Key words
floating wind power platform /
deep learning /
bi-directional long short-term memory（Bi-LSTM） /
motion prediction /
neural network /
wave load
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 何鸿圣, 李春, 王博, 等. 2种海上风力机漂浮式风电场平台动态响应对比[J]. 太阳能学报, 2023, 44(4): 1-8.HE H S, LI C, WANG B, et al. Comparison of dynamic response of two floating wind farm platforms for offshore wind turbines[J]. Acta energiae solaris sinica, 2023, 44(4): 1-8.
[2] 王博, 李春, 丁勤卫, 等. 基于正交化的混合式平台漂浮式风电场平台动态响应研究[J]. 太阳能学报, 2022, 43(1): 80-88.WANG B, LI C, DING Q W, et al. Research on dynamic response of hybrid platform floating wind farm platform based on orthogonalization[J]. Acta energiae solaris sinica, 2022, 43(1): 80-88.
[3] KHAN A, BIL C, MARION K E.Ship motion prediction for launch and recovery of air vehicles[C]//Proceedings of OCEANS 2005 MTS/IEEE. Washington, DC, USA, 2006: 2795-2801.
[4] BREMER K S.Using neural networks to predict the response of a floating structure[D]. Trondheim, Norwegian University of Science and Technologt, 2018.
[5] MAZAHERI S.The usage of artificial neural networks in hydrodynamic analysis of floating offshore platforms[J]. The international journal of maritime engineering, 2006, 3(4): 48-60.
[6] SHENG W, TONY L.Artificial neural network application in short-term prediction in an oscillating water column[C]//ISOPE International Ocean and Polar Engineering Conference. Beijing, China, 2010.
[7] 牟哲岳, 孙勇, 王瑞良, 等. 基于实测数据和机器学习的风电机组载荷预测模型[J]. 太阳能学报, 2023, 44(10): 414-419.MOU Z Y, SUN Y, WANG R L, et al. Prediction model for wind turbine loads based on experimental data and machine learning[J]. Acta energiae solaris sinica, 2023, 44(10): 414-419.
[8] 庞军恒, 黄炜楠, 董胜. 基于多变量DSD-LSTM模型的有效波高预测[J]. 太阳能学报, 2024, 45(7): 121-127.
PANG J H, HUANG W N, DONG S.Prediction of significant wave height based on multivariable DSD-LSTM model[J]. Acta energiae solaris sinica, 2024, 45(7): 121-127.
[9] 魏强强. 基于深度学习的半潜平台运动响应分析与预测[D]. 上海: 上海交通大学, 2021.
WEI Q Q.Analysis and prediction of semi-submersible platform motion based on deep learning[D]. Shanghai: Shanghai Jiao Tong University, 2021.
[10] GUO X X, ZHANG X T, TIAN X L, et al.Predicting heave and surge motions of a semi-submersible with neural networks[J]. Applied ocean research, 2021, 112: 102708.
[11] 陈海, 李志刚, 冯加果. 基于深度置信神经网络的半潜式平台浮体运动模型和响应预测研究[J]. 船舶力学, 2021, 25(5): 586-597.CHEN H, LI Z G, FENG J G. Prediction of motion and its distribution laws of semi-subsea platform based on deep learning methods[J]. Journal of ship mechanics, 2021, 25(5): 586-597.
[12] GUO X X, ZHANG X T, TIAN X L, et al.Probabilistic prediction of the heave motions of a semi-submersible by a deep learning model[J]. Ocean engineering, 2022, 247: 110578.
[13] 姚骥. 基于深度学习的海洋环境、浮体响应预测及水下结构异常诊断[D]. 大连: 大连理工大学, 2022.
YAO J.Ocean environment, floating body response prediction and abnormal diagnosis of underwater structures based on deep learning method[D]. Dalian: Dalian University of Technology, 2022.
[14] DENG Y F, FENG W, XU S W, et al.A novel approach for motion predictions of a semi-submersible platform with neural network[J]. Journal of marine science and technology, 2021, 26(3): 883-895.
[15] HOCHREITER S, SCHMIDHUBER J.Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[16] DYER C, BALLESTEROS M, LING W, et al.Transition-based dependency parsing with stack long short-term memory[J]. Computer science, 2015, 37(2): 321-332.
[17] YU W, KOLJA M, FRANK L.D4.2 public definition of the two LIFES50+10 MW floater concepts[R]. Stuttgart: University of Stuttgart, 2018.
基金
国家重点研发计划（2023YFE0102000）; 中国长江三峡集团有限公司科研项目（202303059）; 国家自然科学基金（52476205）; 高等学校学科创新引智计划（111计划）“跨海大桥安全保障与智能运行学科创新引智基地(D21013)”
PDF(1596 KB)
