| 48 | 0 | 43 |
| 下载次数 | 被引频次 | 阅读次数 |
[目的]为满足漂浮式风电平台运动与系泊系统张力的短期预测需求，且不依赖实时运动数据，[方法]提出一种环境参数驱动的深度学习预测方法。采用随机环境采样和精细化数值仿真构建训练数据集，设计双分支全连接网络分别学习运动与系泊系统张力特征。模型仅以环境参数为输入，输出六自由度运动极值与系泊系统张力极值，采用Z-score标准化、Dropout正则化及Adam优化器提升泛化能力。[结果]在国能共享号平台案例中，除艏摇外，其余五自由度运动预测的拟合优度R2>0.95;7根系泊缆的张力预测R2>0.98，最大平均绝对误差为75.29 kN，占预张力的11%，精度满足工程要求。[结论]研究成果可为漂浮式风电运维安全决策提供一定参考。
Abstract：[Purpose] To meet the demand for short term forecasting of floating wind turbine platform motions and mooring forces while eliminating reliance on real-time motion data, [Method] a deep learning forecasting method driven by environmental parameters is proposed. Training datasets are constructed using random environmental sampling and refined numerical simulations. A dual-branch fully connected network is designed to learn motion and mooring force features separately. The model takes only environmental parameters as input and outputs six-degree-of-freedom motion extremes and mooring force extremes. Z-score normalization, Dropout regularization, and the Adam optimizer are applied to enhance generalization capability. [Result] In the case study of the "Guo Neng Gong Xiang Hao" platform, R2 exceeded 0.95 for all motion degrees of freedom except yaw. For mooring force prediction, R2 is greater than 0.98 for seven lines, with a maximum mean absolute error of 75.29 k N, accounting for 11% of the pretension, meeting engineering accuracy requirements. [Conclusion] The research results can provide some references for the safety decision-making of floating wind power operation and maintenance.
[1]LI H, DIAZ H, SOARES C G. A Developed Failure Mode and Effect Analysis for Floating Offshore Wind Turbine Support Structures[J]. Renewable Energy, 2021,164:133-145.
[2]陈政权,刘昆,宋娜,等.风-浪-系泊耦合的浮式风机在碰撞载荷下的动力响应[J].船舶工程, 2025, 47(8):152-166.CHEN Z Q, LIU K, SONG N, et al. Dynamic Response of Floating Wind Turbines Under Collision Loads with Wind-Wave-Moor Coupling[J]. Ship Engineering, 2025,47(8):152-166.
[3]戴仰山,沈进威,宋竟正.船舶波浪荷载[M].北京:国防工业出版社, 2007.DAI Y S, SHEN J W, SONG J Z. Ship Wave Loads[M].Beijing:National Defense Industry Press, 2007.
[4]WIENER N. Extrapolation, Interpolation and Smoothing of Stationary Time Series, with Engineering Applications[J]. Journal of the Royal Statal Society Series A, 1950, 113(3):413-421.
[5]SIDAR M, DOOLIN B. On the Feasibility of Real-Time Prediction of Aircraft Carrier Motion at Sea[J]. IEEE Transactions on Automatic Control, 1983, 28(3):350-356.
[6]EHOCHSON L D. Prediction of Time Series Using Multiple Regression Techniques and Seakeeping Applications[J]. Prediction of Time, 1963, 33:1-48.
[7]YUMORI I R. Real Time Prediction of Ship Response to Ocean Waves Using Time Series Analysis[C]//OCEANS 1981. 1981.
[8]KHAN A, BIL C, MARION K E. Ship Motion Prediction for Launch and Recovery of Air Vehicles[J].OCEANS 2005. 2005.
[9]ROBERTS J B, SPANOS P D. Stochastic Averaging:an Approximate Method of Solving Random Vibration Problems[J]. International Journal of Non-Linear Mechanics, 1986, 21(2):111-134.
[10]DOSTAL L, KREUZER E, SØRENSEN A J. Stochastic Analysis of Mooring Systems Using the Complex Fractional Derivative Model[C]//ASME. 2012.
[11]MATSUI S, OKA M. Long-Term Prediction of Nonlinear Ship Roll Motion Using RAO[J]. Applied Ocean Research, 2025, 158:104590.
[12]CHAI W, NAESS A, LERIA B J. Stochastic Response Analysis of Mooring Cables via Path Integration[R].2016.
[13]MAKI A, WANG D, SOLTANI O M, et al. Line Integral Method for the Long-Term Analysis of Mooring Systems[J]. Journal of Offshore Mechanics and Arctic Engineering, 2021, 143(1):011701.
[14]MATSUI S, OKA M. RAO-Based Translation Process for Efficient Prediction of Nonlinear Vessel Motions[C]//33rd International Conference on Offshore Mechanics and Arctic Engineering. 2023.
[15]BJØRNI F A, LIEN S, MIDTGARDEN T A, et al.Prediction of Dynamic Mooring Responses of a Floating Wind Turbine Using an Artificial Neural Network[J].IOP Conference Series. Materials Science and Engineering, 2021, 1201(1):012023.
[16]WANG Z, QIAO D, YAN J, et al. A New Approach to Predict Dynamic Mooring Tension Using LSTM Neural Network Based on Responses of Floating Structure[J].Ocean Engineering, 2022, 249:110905.
[17]胡乐涵.基于长短期记忆神经网络的漂浮式海上风机动力响应研究[D].辽宁大连:大连理工大学, 2024.HU L H. Dynamic Response Study of Floating Offshore Wind Turbine Based on Long-Short Term Memory Neural Networks[D]. Dalian, Liaoning:Dalian University of Technology, 2024.
[18]孙刘璐,于思源,武文华,等.基于现场实测数据的半潜式平台运动响应分析和极值预测研究[J].海洋工程, 2022, 40(1):57-64.SUN L L, YU S Y, WU W H, et al. Motion Analysis and Extreme Value Prediction of Semi-Submersible Platform Based on Prototype Monitoring Data[J]. Ocean Engineering, 2022, 40(1):57-64.
[19]赵大文,张黎.基于一体化仿真的几种典型风电半潜漂浮式基础的对比研究[J].能源工程, 2025, 45(3):83-95.ZHAO D W, ZHANG L. Comparative Study on Typical Semi-submerged Floating Foundations for Wind Turbine Based on Integrated Simulation[J]. Energy Engineering, 2025, 45(3):83-95.
基本信息:
DOI：10.13788/j.cnki.cbgc.2026.04.02
中图分类号:TM614;TP18
引用信息:
[1]李红有,樊健生,周全智,等.环境参数驱动的漂浮式风电平台运动与系泊系统张力预测[J].船舶工程,2026,48(04):9-19.DOI:10.13788/j.cnki.cbgc.2026.04.02.
基金信息:
国家自然科学基金项目(52121005); 龙源电力集团股份有限公司科技项目(LYX-2024-07)
2026-04-25
2026-04-25