提出一种基于深度神经网络的风力机疲劳载荷代理模型,旨在解决风力机场址评估效率低下的问题。开展基于深度神经网络（DNN）的风力机疲劳载荷代理模型研究。首先,根据平均风速、湍流强度、风切变、偏航误差、入流角和空气密度6维环境变量的分布和相关性进行准蒙特卡洛抽样,获得10000个环境变量样本。然后,采用TurbSim和OpenFAST对NREL 5 MW参考风力机进行仿真得到载荷时程,并通过MLife计算得到1 Hz的等效疲劳载荷（DEL）数据库。最后,运用DNN方法建立DEL的代理模型并对模型精度进行详细验证。结果表明：基于DNN的DEL代理模型具有较高的预测精度,计算效率得到显著提升。
Abstract
A deep neural network （DNN）-based surrogate model for wind turbine fatigue load is proposed to address the low-efficiency of site-specific suitability assessment. A surrogate model of fatigue loads of wind turbines based on deep neural network （DNN） is comprehensively investigated. Firstly, 10000 samples of six environmental variable space including average wind speed, turbulence intensity, wind shear, yaw misalignment, vertical inflow angle and air density are generated by quasi-Monte Carlo method. Then, TurbSim and OpenFAST are used to simulate the load time history of the NREL 5MW reference wind turbine, and MLife is used to obtain the damage equivalent load （DEL） database at 1 Hz. Finally, the DNN method is used to establish DEL surrogate models for 7 load channels, and the accuracy of the models are extensively verified. Results show that the DNN-based DEL surrogate models have high prediction accuracy and computational efficiency.
关键词
风力机 /
疲劳载荷 /
OpenFAST /
深度神经网络 /
代理模型
Key words
wind turbine /
fatigue load /
OpenFAST /
deep neural network /
surrogate model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 胡春成. 风力发电发展现状以及行业发展浅谈[J]. 科技创新与生产力, 2023(3): 55-57.
HU C C.Brief discussion on the wind power generation development status and industry development[J]. Sci-tech innovation and productivity, 2023(3): 55-57.
[2] IEC61400-1-2019, Wind turbine generator systems- part 1: safety requirements[S].
[3] GANESH N, GUPTA S.Estimating the rain-flow fatigue damage in wind turbine blades using polynomial chaos[C]//Proceedings of the International Symposium on Engineering under Uncertainty: Safety Assessment and Management (ISEUSAM - 2012), India: Springer India, 2012: 711-727.
[4] TOFT H S, SVENNINGSEN L, MOSER W, et al.Assessment of wind turbine structural integrity using response surface methodology[J]. Engineering structures, 2016, 106: 471-483.
[5] DIMITROV N, KELLY M C, VIGNAROLI A, et al.From wind to loads: wind turbine site-specific load estimation with surrogate models trained on high-fidelity load databases[J]. Wind energy science, 2018, 3(2): 767-790.
[6] 付瑜, 陈鹏, 邹亮, 等. 基于数据库的风电场疲劳载荷适应性平台[J]. 船舶工程, 2020, 42(增刊2): 269-272.
FU Y,CHEN P,ZOU L,et al.Wind farm fatigue load site suitability tool based on database[J]. Ship engineering, 2020, 42(S2): 269-272.
[7] 谢炜. 基于环境参数的风电机组疲劳载荷代理模型研究[J]. 风能, 2022(5): 76-83.
XIE W.Research on fatigue load proxy model of wind turbine based on environmental parameters[J]. Wind energy, 2022(5): 76-83.
[8] ANANTHAN S, VIJAYAKUMAR G, YELLAPANTULA S.A DNN surrogate unsteady aerodynamic model for wind turbine loads calculations[J]. Journal of physics: conference series, 2020, 1618(5): 052060.
[9] YAO Q, MA B, ZHAO T Y, et al.Optimized active power dispatching of wind farms considering data-driven fatigue load suppression[J]. IEEE transactions on sustainable energy, 2023, 14(1): 371-380.
[10] 付德义, 张晓东, 王瑞明, 等. 特定场址条件下风电机组载荷适应性评估[J]. 太阳能学报, 2021, 42(6): 425-431.
FU D Y, ZHANG X D, WANG R M, et al.Wind turbine load adaptability assessment under specific site conditions[J]. Acta energiae solaris sinica, 2021, 42(6): 425-431.
[11] 林剑威, 沈文忠. 考虑疲劳载荷的风电场布局优化[C]//中国可再生能源学术大会-风能论文集, 南京, 中国, 2023.
LIN J W, SHEN W Z.Optimization of wind farm layout considering fatigue load[C]//Proceedings of China Renewable Energy Society-Wind Energy. Nanjing, China, 2023.
[12] CAFLISCH R E.Monte Carlo and quasi-Monte Carlo methods[J]. Acta numerica, 1998, 7: 1-49.
[13] 张箭. 高海拔山区风电场风电机组选型探讨[J]. 云南水力发电, 2018, 34(5): 13-16.
ZHANG J.Discussion on type selection of A wind farm turbine unit in a high altitude mountainous region[J]. Yunnan water power, 2018, 34(5): 13-16.
[14] JONKMAN B J,BUHL M L.TurbSim user’s guide for version 140[R]. National Renewable Energy Laboratory, Technical Report Golden, Colorado, 2008.
[15] HAYMAN G J, BUHL J M.Mlife users guide for version 1.00[J]. National renewable energy laboratory, 2012, 74(75): 112.
[16] 韩山杰, 谈世哲. 基于TensorFlow进行股票预测的深度学习模型的设计与实现[J]. 计算机应用与软件, 2018, 35(6): 267-271, 291.
HAN S J, TAN S Z.Design and implementation of deep learning model for stock forecasting based on TensorFlow[J]. Computer applications and software, 2018, 35(6): 267-271, 291.
[17] ABADI M, BARHAM P, CHEN J M, et al.TensorFlow: a system for large-scale machine learning[J]. Proceedings of the 12th USENIX symposium on operating systems design and implementation, OSDI 2016, 2016: 265-283.
[18] SCHRÖDER L, KRASIMIROV DIMITROV N, VERELST D R, et al. Wind turbine site-specific load estimation using artificial neural networks calibrated by means of high-fidelity load simulations[J]. Journal of physics: conference series, 2018, 1037: 062027.
[19] 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.
[20] 韩耀振, 刘爽, 杨文祥, 等. 基于模糊LQR的Spar型海上浮式风力机载荷抑制[J]. 太阳能学报, 2024, 45(1): 188-196.
HAN Y Z, LIU S, YANG W X, et al.Load suppression of Spar-types offshore floating wind turbine based on fuzzy LQR[J]. Acta energiae solaris sinica, 2024, 45(1): 188-196.
基金
国家自然科学基金面上项目（52378480）; 重庆市自然科学基金创新发展联合基金（CSTB2024NSCQ-LZX0010）; 学科创新引智基地项目（B18062）; 中央高校基本科研业务费项目（2022CDJQY-009）
PDF(1990 KB)
