焦晓峰,蒋兴群,刘波,宋力,陈永艳,张宪琦
JIAO Xiaofeng,JIANG Xingqun,LIU Bo,SONG Li,CHEN Yongyan,ZHANG Xianqi

• 1:内蒙古电力科学研究院
• 2:内蒙古工业大学
• 3:内蒙古自治区可再生能源重点实验室
• 4:哈尔滨工业大学计算学部

摘要(Abstract)：

为了对风力发电机组叶片损伤状态进行有效检测，提出一种基于CenterNet目标检测算法的风机叶片损伤检测识别技术。该技术选取DLA-60特征提取网络作为CenterNet算法的骨干网络，并在DLA-60网络中引入注意力引导数据增强机制，提升检测算法的精度。优化后风力机叶片损伤检测识别模型的检测精度为88%，较原始算法提升了2.6个百分点，且检测时间基本与原网络持平，具有较强的精确性和实用性。
In order to effectively detect the damage state of wind turbine blades, this paper proposes a wind turbine blade damage detection and recognition technology based on the CenterNet target detection algorithm. The DLA-60 feature extraction network is selected as the backbone network of the CenterNet algorithm, and the attention-guided data enhancement mechanism is introduced in the DLA-60 network to improve the accuracy of the detection algorithm.Experiments show that the optimized wind turbine blade damage detection and recognition model has a detection accuracy of 88%, which is 2.6% higher than the original algorithm, and the detection time is basically the same as the original network, which has great practicability and accuracy.

关键词(KeyWords)： 风力机叶片;损伤检测;CenterNet算法;Attention机制;卷积神经网络
turbine blade;damage detection;CenterNet algorithm;Attention mechanism;convolutional neural network

Abstract:

Keywords:

基金项目(Foundation): 内蒙古自治区2019年科技项目“风力机叶片结构动态响应研究及裂纹检测应用示范”

作者(Author): 焦晓峰,蒋兴群,刘波,宋力,陈永艳,张宪琦
JIAO Xiaofeng,JIANG Xingqun,LIU Bo,SONG Li,CHEN Yongyan,ZHANG Xianqi

参考文献(References)：

• [1] A. Tummala, R. K. Velamati, D.K. Sinha, et al. A review on small scale wind turbines[J]. Renewable Sustainable Enery Reviews, 2016, 152(56):1351-1371.
• [2] H. D. M. de Azevedo, A. M. Araújo, N.Bouchonneau.A review of wind turbine bearing condition monitoring:state of the art and challenges[J]. Renewable Sustain able Enery Reviews, 2016, 152(56):368-379.
• [3] CHEN X, YAN R, LIU Y. Wind turbine condition monitoring and fault diagnosis in china[J]. IEEE Instrum. Meas. Mag., 2016, 19(2):22-28.
• [4] TIAN Kangqi, SONG Li, JIAO Xiaofeng, et al. Aeroelastic stability analysis and failure damage evaluation of wind turbine blades under variable conditions[J/OL].Energy Sources, 2021:1-18[2021-03-25]. https://www.tandfonline.com/doi/full/10.1080/15567036.2021.1924320.
• [5]曾军，陈艳峰，杨苹，等.大型风力发电机组故障诊断综述[J].电网技术，2018,42(3):849-860.ZENG Jun, CHEN Yanfeng, Yang Ping, et al. Review of fault diagnosis methods of large-scale wind turbines[J]. Power System Technology, 2018, 42(3):849-860.
• [6]李刚，齐莹，李银强，等.风力发电机组故障诊断与状态预测的研究进展[J].电力系统自动化，2021,45(4):180-191.LI Gang, QI Ying, LI Yinqiang, et al. Research Progress on fault diagnosis and state prediction of wind turbine[J]. Automation of Electric Power Systems, 2021, 45(4):180-191.
• [7]陈雪峰，郭艳婕，许才彬，等.风电装备故障诊断与健康监测研究综述[J].中国机械工程，2020,31(2):175-189.CHEN Xuefeng, GUO Yanjie, XU Caibin, et al. Review of fault diagnosis and health monitoring for wind power equipment[J]. China Mechanical Engineering,2020, 31(2):175-189.
• [8]杨昌杏，李录平.风力机叶片损伤故障检测技术研究进展[J].发电技术，2020,41(6):599-607.YANG Changxing, LI Luping. Research progress on blade damage fault detection technology of wind turbine[J]. Power Generation Technology, 2020, 41(6):599-607.
• [9]范晨亮，李国庆，马长啸，等.基于深度学习的风机叶片裂纹检测算法[J].科学技术创新，2020(13):72-75.FAN Chenliang, LI Guoqing, MA Changxiao, et al.Fan blade crack detection algorithm based on deep learning[J]. Science and Technology Innovation, 2020(13):72-75.
• [10] WANG Long. Automatic detection of wind turbine blade surface cracks based on UAV-taken images[J].IEEE Transactions on Industrial Electronics, 2017, 64(9):7293-7303.
• [11] ZHOU X. Y., WANG D. Q., P. Kr?henbühl. Objects as points[J/OL]. Cornell University Library, Ithaca,NY, 2019. https://arxiv.org/abs/1904.07850.
• [12] YU F., WNAG D., E. Shelhamer E, et al. Deep layer aggregation[J/OL]. Cornell University Library,Ithaca, NY, 2017. https://arxiv.org/abs/1707.06484.
• [13] HU T, QI H, HUANG Q, et al. See better before look ing closer:Weakly supervised data augmentation net work for fine-grained visual classification[J/OL].Cornell University Library, Ithaca, NY, 2019. https://arxiv.org/abs/1901.09891v1.
• [14]杨家欢，宗哲英，王祯，等.风机叶片检测的研究现状及进展[J].复合材料科学与工程，2020(6):109-113.YANG Jiahuan, ZONG Zheying, WANG Zhen, et al.A review of research on detecting the wind turbine blades[J]. Composites Science and Engineering, 2020(6):109-113.
• [15]杜铁峰.300 MW循环流化床锅炉双级动叶引风机常见故障原因分析及建议[J].机电工程技术，2020,49(5):216-219.DU Tiefeng. Analysis and suggestion on common faults of double stage moving blade induced draft fan of 300 MW CFB boiler[J]. Mechanical&Electrical Engineering Technology, 2020, 49(5):216-219.
• [16]余金泰.风力发电机组故障诊断与预测技术研究综述[J].中国科技纵横，2021(13):95-96.YU Jintai. Summary of research on fault diagnosis and prediction technology of wind turbine[J]. China Science&Technology Overview, 2021(13):95-96.
• [17]柯昭，付育良，黄锴.引风机叶片磨损故障诊断及原因分析[J].东北电力技术，2019,40(11):35-37,62.KE Zhao, FU Yuliang, HUANG Kai. Analysis of induced draft fan wear fault diagnosis and reason[J].Northeast Electric Power Technology, 2019, 40(11):35-37, 62.