基于续尺度卷积网络的10MW漂浮式风力机筋腱损伤识别

PDF(3446 KB)

振动与冲击 ›› 2022, Vol. 41 ›› Issue (3) : 183-189.

论文

许子非1,2，杨阳2,3，李春1,4，缪维跑1，张万福1，金江涛1，王鑫雨1

作者信息 +

Tendon damage identification of 10 MW floating wind turbine based on CMS-CNN

XU Zifei1,2, YANG Yang2,3, LI Chun1,4, MIAO Weipao1, ZHANG Wanfu1, JIN Jiangtao1, WANG Xinyu1

Author information +

文章历史 +

摘要

为提升复杂环境中漂浮式风力机平台筋腱结构隐性损伤识别率，基于卷积神经网络(Convolutional Neural Network，CNN)，提出连续多尺度卷积神经网络（Continues-Multi-Scale Convolutional Neural Network，CMS-CNN），建立“端到端”的损伤识别模型。为验证CMS-CNN方法的有效性，以10MW漂浮式风力机为研究对象，对损伤位置、程度进行故障诊断，结果表明：连续多尺度模型比传统多尺度的诊断结果更佳；横荡加速度受环境载荷影响较小，基于此响应信号所训练的CMS-CNN诊断模型更可靠；CMS-CNN模型可在筋腱结构微弱损伤时实现精准定位，亦能完成结构隐性损伤程度识别。

Abstract

A novel end-to-end Diagnosis model named Continuous-Multi-Scale Convolutional Neural Network (CMS-CNN) has been proposed to improve the rate of identification of structural damage on the damaged tendons in the float wind turbine platform. The effectiveness of the proposed CMS-CNN is examined by a 10MW float wind turbine model. The damaged locations and damaged degrees of the tendons in the 10MW float wind turbine are diagnosed by the proposed method. The results show that: The model considered more information by the continuous multi-scale coarse-grained procedure has better performance than the traditional multi-scale based model. The CMS-CNN model, using sway acceleration as the inputs, is more reliable than the model using the other accredited information. The CMS-CNN model can locate the damaged positions in the initial stages of damage evolution and diagnose both the locations and degrees of recessive damage of the tendons.

导出引用

许子非1,2，杨阳2,3，李春1,4，缪维跑1，张万福1，金江涛1，王鑫雨1. 基于续尺度卷积网络的10MW漂浮式风力机筋腱损伤识别[J]. 振动与冲击, 2022, 41(3): 183-189

XU Zifei1,2, YANG Yang2,3, LI Chun1,4, MIAO Weipao1, ZHANG Wanfu1, JIN Jiangtao1, WANG Xinyu1. Tendon damage identification of 10 MW floating wind turbine based on CMS-CNN[J]. Journal of Vibration and Shock, 2022, 41(3): 183-189

参考文献

[1] 迟永宁，梁伟，张占奎，等．大规模海上风电输电与并网关键技术研究综述[J]．中国电机工程学报，2016，36(14)：3758-3771．
CHI Yongning，LIANG Wei，ZHANG Zhankui, et al. An Overview on Key Technologies Regarding Power Transmission and Grid Integration of Large Scale Offshore Wind Power[J]. Proceedings of the CSEE, 2016, 36(14): 3758-3771.
[2] 黄致谦，丁勤卫，李春，等．新型漂浮式风力机半潜平台抑制摇荡运动设计研究[J]．中国电机工程报，2018，38(24)：7295-7303+7456．
HUANG Zhiqian，DING Qinwei，LI Chun, et al. Design and Research on Suppression Swaying Motion of the New Semi-submersible Platform of Floating Wind Turbine[J]. Proceedings of the CSEE, 2018, 38(24): 7295-7303+7456.
[3] 方龙，李良碧，李荣富．海上漂浮式风力机支撑结构疲劳寿命研究[J]．太阳能学报，2016，37(12)：3184-3188．
FANG Long，LI Liangbi，LI Rongfu. Fatigue life research of support structure for floating offshore wind turbine[J]. Acta Energiae Solaris Sinica, 2016, 37(12): 3184-3188.
[4] 金鑫，钟翔，何玉林，等．漂浮特性对风力发电机振动影响[J]．振动与冲击，2013(15)：33-38．
JIN Xin，ZHONG Xiang，HE Yulin et al.Floating characteristics' impacton vibration of wind turbine [J]. Journal of Sound and Vibration. 32(15): 26-31.
[5] Yang Y, Musa B, Jin W, et al. Wind-wave coupling effects on the fatigue damage of tendons for a 10MW multi-body floating wind turbine[J]. Ocean Engineering,2020, 217: 107909.
[6] 叶舟，成欣，周国龙，等．漂浮式风力机平台动态响应的优化方法探讨[J]．振动与冲击, 2016，000(003)：20-26.
YE Zhou，CHENG Xin，ZHOU Guolong, et al. Optimization method for dynamic responses of floating offshore wind turbines [J]. Journal of Sound and Vibration. 2016, 000(003):20-26. [7] Ettefagh M M . Damage Identification of A TLP Floating Wind Turbine by Meta-Heuristic Algorithms [J]. China Ocean Engineering, 2015, 29(6):891-902.
[8] Begg S E , Fowkes N , Stemler T , et al. Fault detection in vibration systems: Identifying damaged moorings[J]. Ocean Engineering, 2018, 164(15):577-589.
[9] JAHANGIRI V , FATHI R, MOHAMMAD ETTEFAGH M, et al. TLP Structural Health Monitoring Based on Vibration Signal of Energy Harvesting System[J]. Latin American Journal of Solids & Structures, 2016, 13(5):897-915.
[10] Chandrasekaran S , Chithambaram T . Health monitoring of tension leg platform using wireless sensor networking: experimental investigations[J]. Journal of Marine Science and Technology, 2019, 24:60-72.
[11] MOHAMMAD A L Y , SHAHVERDI S, TARINEJAD R, et al. Structural Health Monitoring (SHM) of Offshore Jacket Platforms[C]// ASME 2011 30th International Conference on Ocean, Offshore and Arctic Engineering. ASME , Rotterdam, 2011.
[12] Jiang Q , Li T , Yao Y , et al. A method of characteristic impulse identification based on Laplace wavelet matching[C]// 2010 3rd International Congress on Image and Signal Processing. Yantai :IEEE, 2010.
[13] Hakim S J S , Razak H A . Modal parameters based structural damage detection using artificial neural networks - a review[J]. Smart structures and systems, 2014, 14(2):159-189.
[14] 陈保家，黄伟，李立军，等．最大重叠离散小波包变换边际谱特征在齿轮故障诊断中的应用[J]．西安交通大学学报，2020，54(02)：35-42．

CHEN Baojia,HUANG Wei,LI Lijun, et al. Application of Maximum Overlap Discrete Wavelet Packet Transform Marginal Spectrum Features in Gear Fault Diagnosis[J]. Journal of Xi'an Jiaotong University, 2020, 54(02): 35-42.
[15] Yang Y , Bashir M , Li C , et al. Investigation on mooring breakage effects of a 5 MW barge-type floating offshore wind turbine using F2A[J]. Ocean Engineering, 2021, 233(1): 108887
[16] Luo Y, Cheng Y, UZUNER Ö, et al. Segment convolutional neural networks (Seg-CNNs) for classifying relations in clinical notes[J]. Journal of the American Medical Informatics Association, 2017, 25(1):93-98.
[17] Kuo C C J. Understanding Convolutional Neural Networks with A Mathematical Model [J]. Journal of Visual Communication & Image Representation, 2016, 41:S1047320316302267.
[18] Jiang G Q, He H B, Yan J, et al. Multiscale Convolutional Neural Networks for Fault Diagnosis of Wind Turbine Gearbox [J] IEEE transaction on Industrial Electronics, 2019,66,(4):3196-3207.