基于迁移学习的风电机组叶片损伤检测与分析

殷孝雎; 潘雪; 左雁斌; 关新

doi:10.19912/j.0254-0096.tynxb.2023-0925

PDF(1599 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (10) : 506-511. DOI: 10.19912/j.0254-0096.tynxb.2023-0925

基于迁移学习的风电机组叶片损伤检测与分析

殷孝雎¹, 潘雪¹, 左雁斌², 关新¹

作者信息 +

WIND TURBINE BLADE DAMAGE DETECTION AND ANALYSIS BASED ON TRANSFER LEARNING

Yin Xiaoju¹, Pan Xue¹, Zuo Yanbin², Guan Xin¹

Author information +

文章历史 +

摘要

针对风电机组叶片损伤成因复杂、故障识别效率低、精度不足等问题,提出一种基于迁移学习改进的DenseNet网络（DenseNet-TL）的风电机组叶片损伤检测方法。建立DenseNet-TL数学模型,提升特征提取能力,在该模型下对风电机组叶片图像进行识别分析,以确定叶片的损伤状态。以某风场数据集进行离线训练和测试,结果表明：与AlexNet、ResNet模型进行对比,该模型可有效节省训练时间、提高模型的泛化能力,训练准确度平均值达到90%以上,验证了该方法的有效性和精确性。

Abstract

Aiming at the problems of complex causes of wind turbine blade damage, low fault identification efficiency, and insufficient accuracy, a wind turbine blade damage detection method is propsed based on improved DenseNet network improved by transfer learning is proposed. A mathematical model of DenseNet network improved by transfer learning （DenseNet-TL ） is established to improve the feature extraction capability, and the recognition and analysis of wind turbine blade images are carried out under the model to determine the damage state of the blades. The offline training and testing is carried out with a wind farm dataset, and the results show that, compared with AlexNet and ResNet models, the model effectively saves the training time, improves the generalization ability of the model, and the average training accuracy reaches more than 90%, which verifies the validity and accuracy of the method.

导出引用

殷孝雎, 潘雪, 左雁斌, 关新. 基于迁移学习的风电机组叶片损伤检测与分析[J]. 太阳能学报. 2024, 45(10): 506-511 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0925

Yin Xiaoju, Pan Xue, Zuo Yanbin, Guan Xin. WIND TURBINE BLADE DAMAGE DETECTION AND ANALYSIS BASED ON TRANSFER LEARNING[J]. Acta Energiae Solaris Sinica. 2024, 45(10): 506-511 https://doi.org/10.19912/j.0254-0096.tynxb.2023-0925

中图分类号： TK83

参考文献

[1] 覃盛琼, 程朗, 何占启, 等. 风力发电系统研究与应用前景综述[J]. 机械设计, 2021, 38(8): 1-8.
QIN S Q, CHENG L, HE Z Q, et al.Review of research and application on the wind power-generation system[J]. Journal of machine design, 2021, 38(8): 1-8.
[2] 刘双, 胡斌, 贺铸, 等. 风机叶片缺陷的无损检测方法比较与评价[J]. 无损检测, 2015, 37(3): 47-51.
LIU S, HU B, HE Z, et al.Comparison and evaluation on the non-destructive testing methods for the wind turbine blades defects[J]. Nondestructive testing, 2015, 37(3): 47-51.
[3] 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 J H, ZONG Z Y, WANG Z, et al.A review of research on detecting the wind turbine blades[J]. Composites science and engineering, 2020(6): 109-113.
[4] 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 C X, LI L P.Research progress on blade damage fault detection technology of wind turbine[J]. Power generation technology, 2020, 41(6): 599-607.
[5] 汤占军, 肖遥, 李英娜. 基于深度学习的风机叶片结冰诊断方法[J]. 现代电子技术, 2022, 45(11): 77-82.
TANG Z J, XIAO Y, LI Y N.Wind turbine blade icing diagnosis method based on deep learning[J]. Modern electronics technique, 2022, 45(11): 77-82.
[6] BI J X, FAN W Z, WANG Y, et al.A fault diagnosis algorithm for wind turbine blades based on BP neural network[J]. IOP conference series: materials science and engineering, 2021, 1043(2): 022032.
[7] OU Y W, CHATZI E N, DERTIMANIS V K, et al.Vibration-based experimental damage detection of a small-scale wind turbine blade[J]. Structural health monitoring, 2017, 16(1): 79-96.
[8] LIU Z P, ZHANG L, CARRASCO J.Vibration analysis for large-scale wind turbine blade bearing fault detection with an empirical wavelet thresholding method[J]. Renewable energy, 2020, 146: 99-110.
[9] 焦晓峰, 蒋兴群, 刘波, 等. 基于改进CenterNet算法的风机叶片损伤检测识别技术[J]. 内蒙古电力技术, 2022, 40(1): 10-14.
JIAO X F, JIANG X Q, LIU B, et al.Wind turbine blade damage detection recognition technology based on improved CenterNet algorithm[J]. Inner Mongolia electric power, 2022, 40(1): 10-14.
[10] 顾永强, 冯锦飞, 张哲玮, 等. 基于模态参数的在役风力发电机叶片损伤识别研究[J]. 太阳能学报, 2022, 43(3): 350-355.
GU Y Q, FENG J F, ZHANG Z W, et al.Research on blade damage identification of active wind turbine based on modal parameters[J]. Acta energiae solaris sinica, 2022, 43(3): 350-355.
[11] MOVSESSIAN A, CAVA D G, TCHERNIAK D.An artificial neural network methodology for damage detection: demonstration on an operating wind turbine blade[J]. Mechanical systems and signal processing, 2021, 159: 107766.
[12] 尹玉, 张永, 王健, 等. 基于热红外图像的风力机叶片损伤识别方法研究[J]. 太阳能学报, 2022, 43(2): 492-497.
YIN Y, ZHANG Y, WANG J, et al.Research on wind turbine blade damage identification method based on thermal infrared image[J]. Acta energiae solaris sinica, 2022, 43(2): 492-497.
[13] 薛刚, 苏天. 基于动力特性及BP神经网络的风电机组叶片损伤识别[J]. 太阳能学报, 2018, 39(10): 2945-2952.
XUE G, SU T.Damage identification of wind turbine blades based on dynamic characteristics and BP neural network[J]. Acta energiae solaris sinica, 2018, 39(10): 2945-2952.
[14] 周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229-1251.
ZHOU F Y, JIN L P, DONG J.Review of convolutional neural network[J]. Chinese journal of computers, 2017, 40(6): 1229-1251.
[15] 林景栋, 吴欣怡, 柴毅, 等. 卷积神经网络结构优化综述[J]. 自动化学报, 2020, 46(1): 24-37.
LIN J D, WU X Y, CHAI Y, et al.Structure optimization of convolutional neural networks: a survey[J]. Acta automatica sinica, 2020, 46(1): 24-37.
[16] 陈超, 齐峰. 卷积神经网络的发展及其在计算机视觉领域中的应用综述[J]. 计算机科学, 2019, 46(3): 63-73.
CHEN C, QI F.Review on development of convolutional neural network and its application in computer vision[J]. Computer science, 2019, 46(3): 63-73.
[17] 吕欢, 许涛, 麻爱松, 等. 基于改进DenseNet和迁移学习的变负载滚动轴承故障诊断[J]. 轻工机械, 2023, 41(1): 53-58.
LYU H, XU T, MA A S, et al.Rolling bearing fault diagnosis method based on improved DenseNet and transfer learning under load changes[J]. Light industry machinery, 2023, 41(1): 53-58.
[18] YANG X Y, ZHANG Y F, LV W, et al.Image recognition of wind turbine blade damage based on a deep learning model with transfer learning and an ensemble learning classifier[J]. Renewable energy, 2021, 163: 386-397.
[19] 董瑞佳. 基于迁移学习和DenseNet的带式输送机托辊故障检测方法[J]. 煤炭技术, 2023, 42(1): 250-252.
DONG R J.Fault detection method of belt conveyor idler based on transfer learning and DenseNet[J]. Coal technology, 2023, 42(1): 250-252.