基于PSO-XGBoost的风电叶片缺陷分类算法

郑浩; 贾展飞; 周丽婷; 王晫

doi:10.19912/j.0254-0096.tynxb.2022-1862

PDF(1472 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (4) : 127-133. DOI: 10.19912/j.0254-0096.tynxb.2022-1862

基于PSO-XGBoost的风电叶片缺陷分类算法

郑浩¹, 贾展飞¹, 周丽婷², 王晫¹

作者信息 +

DEFECT CLASSIFICATION ALGORITHM OF WIND TURBINE BLADES BASED ON PSO-XGBOOST

Zheng Hao¹, Jia Zhanfei¹, Zhou Liting², Wang Zhuo¹

Author information +

文章历史 +

摘要

针对风电叶片缺陷回波信号分类效率低的问题,提出粒子群-极限梯度提升（PSO-XGBoost）算法。首先采用变分模态分解（VMD）结合模糊熵的方法对缺陷回波数据进行特征提取,建立XGBoost多分类模型,在此基础上采用PSO算法对XGBoost超参数进行寻优,建立PSO-XGBoost多分类模型。这种PSO、XGBoost相结合的算法可提高风电叶片缺陷的预测精度、减少缺陷分类的误差。通过仿真,对PSO-XGBoost、XGBoost及其他几种机器学习算法进行对比,结果表明PSO-XGBoost算法准确度最高,其缺陷分类准确率可达98%。由此可见,采用PSO-XGBoost算法可有效提高风电叶片缺陷分类的准确率。

Abstract

To tackle the issue of the low classification efficiency of defect echo signals of wind turbine blades, a particle swarm optimization and extreme gradient boosting algorithm（PSO-XGBoost）is proposed in this paper. Primarily, feature extraction of the defect echo data is applied via variation mode decomposition （VMD） combining with fuzzy entropy to establish XGBoost multiclassification model. Subsequently, XGBoost hyperparameters are optimized via PSO algorithm to establish a PSO-XGBoost multiclassification model. The algorithm of combining of PSO with XGBoost improves the prediction accuracy of wind power blade defects, and reduces the error of defect classification. The simulation results show that PSO-XGBoost algorithm has prediction accuracy by compared with XGBoost and other algorithms, the defects classification accuracy rate of PSO-XGBoost algorithm is up to 98%. Consequently, the PSO-XGBoost algorithm effectively enhances the accuracy of wind turbine blade defect classification.

导出引用

郑浩, 贾展飞, 周丽婷, 王晫. 基于PSO-XGBoost的风电叶片缺陷分类算法[J]. 太阳能学报. 2024, 45(4): 127-133 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1862

Zheng Hao, Jia Zhanfei, Zhou Liting, Wang Zhuo. DEFECT CLASSIFICATION ALGORITHM OF WIND TURBINE BLADES BASED ON PSO-XGBOOST[J]. Acta Energiae Solaris Sinica. 2024, 45(4): 127-133 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1862

中图分类号： TP391

参考文献

[1] 安宗文, 马军霞, 马强, 等. 风电叶片全尺寸疲劳试验加载方法研究[J]. 太阳能学报, 2021, 42(10): 250-257.
AN Z W, MA J X, MA Q, et al.Study on loading scheme for full-scale fatigue test of wind turbine blades[J]. Acta energiae solaris sinica, 2021, 42(10): 250-257.
[2] 贾辉, 张磊安, 王景华, 等. 基于声发射技术的风电叶片复合材料损伤模式识别[J]. 可再生能源, 2022, 40(1): 67-72.
JIA H, ZHANG L A, WANG J H, et al.Damage pattern recognition of wind turbine blade composite material based on acoustic emission technology[J]. Renewable energy resources, 2022, 40(1): 67-72.
[3] DU Y, ZHOU S X, JING X J, et al.Damage detection techniques for wind turbine blades: a review[J]. Mechanical systems and signal processing, 2020, 141: 106445.
[4] HUANG S L, SUN H Y, WANG Q, et al.Unidirectional focusing of horizontally polarized shear elastic waves electromagnetic acoustic transducers for plate inspection[J]. Journal of applied physics, 2019, 125(16): 1-8.
[5] 郑浩, 周丽婷, 王湘明, 等. 电磁超声检测风机叶片缺陷的有限元仿真研究[J]. 电子测量技术, 2021, 44(23): 24-29.
ZHENG H, ZHOU L T, WANG X M, et al.Finite element simulation study on electromagnetic ultrasonic detection of wind turbine blade[J]. Electronic measurement technology, 2021, 44(23): 24-29.
[6] SI D, GAO B, GUO W, et al.Variational mode decomposition linked wavelet method for EMAT denoise with large lift-off effect[J]. NDT & E international, 2019,107: 102149.
[7] 胡松涛, 石文泽, 卢超, 等. 钢轨踏面裂纹电磁超声表面波同步挤压小波快速成像检测研究[J]. 仪器仪表学报, 2020, 41(1): 35-46.
HU S T, SHI W Z, LU C, et al.Research on rapid imaging detection of electromagnetic ultrasonic surface wave synchrosqueezed wavelet for rail tread cracks[J]. Chinese journal of scientific instrument, 2020, 41(1): 35-46.
[8] 李秀艳, 夏琦琦, 王琦, 等. 基于深度学习与多频电磁阵列检测的金属板材表面和内部缺陷识别方法[J]. 电子测量技术, 2021, 44(4): 118-125.
LI X Y, XIA Q Q, WANG Q, et al.A recognition method for sheet metal surface and internal defects based on deep learning and multi-frequency electromagnetic array detection[J]. Electronic measurement technology, 2021, 44(4): 118-125.
[9] 周友行, 孟高磊, 赵文杰, 等. 钢板表面缺陷深度主动学习高效分类方法[J]. 电子测量与仪器学报, 2022, 36(2): 23-31.
ZHOU Y X, MENG G L, ZHAO W J, et al.Efficient deep active learning for steel plate surface defects classification[J]. Journal of electronic measurement and instrumentation, 2022, 36(2): 23-31.
[10] CHEN T Q, GUESTRIN C.XGBoost: a scalable tree boosting system[C]//Proceeding of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA, 2016: 785-794.
[11] TRIZOGLOU P, LIU X L, LIN Z.Fault detection by an ensemble framework of Extreme Gradient Boosting(XGBoost)in the operation of offshore wind turbines[J]. Renewable energy, 2021, 179: 945-962.
[12] 韦海成, 蔡坤, 赵静. 改进VMD算法消除脉搏波基线漂移研究[J]. 电子测量与仪器学报, 2020, 34(8): 144-150.
WEI H C, CAI K, ZHAO J.Study of improved VMD algorithm to eliminate baseline drift of PPG[J]. Journal of electronic measurement and instrumentation, 2020, 34(8): 144-150.
[13] 张勇, 刘洁, 路敬祎, 等. 基于VMD去噪及多尺度模糊熵的管道小泄漏研究[J]. 电子测量技术, 2021, 44(22): 37-43.
ZHANG Y, LIU J, LU J Y, et al.Research on small pipeline leakage based on VMD denoising and multi-scale fuzzy entropy[J]. Electronic measurement technology, 2021, 44(22): 37-43.
[14] NAVANEETH B, SUCHETHA M.PSO optimized 1-D CNN-SVM architecture for real-time detection and classification applications[J]. Computers in biology and medicine, 2019, 108: 85-92.
[15] 居明宇. 基于PSO算法的DBN模型参数优化[J]. 国外电子测量技术, 2020, 39(3): 12-16.
JU M Y.Weight parameter optimization of depth belief network model based on genetic and particle swarm optimization[J]. Foreign electronic measurement technology, 2020, 39(3): 12-16.
[16] DRAGOMIRETSKIY K, ZOSSO D.Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.
[17] KUMAR A, ZHOU Y Q, XIANG J W.Optimization of VMD using kernel-based mutual information for the extraction of weak features to detect bearing defects[J]. Measurement, 2021, 168: 108402.
[18] 杜必强, 孙立江. 变分模态分解和熵理论在超声信号降噪中的应用[J]. 中国工程机械学报, 2017, 15(4): 310-317.
DU B Q, SUN L J.Application of variational mode decomposition and entropy theoryin ultrasonic signal de-noising[J]. Chinese journal of construction machinery, 2017, 15(4): 310-317.
[19] 刘宝生, 邓三鹏, 雷敬. 基于参数优化VMD和模糊熵的自动机故障诊断方法[J]. 机械设计与研究, 2022, 38(2): 93-96, 113.
LIU B S, DENG S P, LEI J.Automaton fault diagnosis method based on VMD with parameter optimization and fuzzy entropy[J]. Machine design & research, 2022, 38(2): 93-96, 113.
[20] LEI Y F, JIANG W L, JIANG A Q, et al.Fault diagnosis method for hydraulic directional valves integrating PCA and XGBoost[J]. Processes, 2019, 7(9): 589.
[21] 吴春华, 董阿龙, 李智华, 等. 基于图相似日和PSO-XGBoost的光伏功率预测[J]. 高电压技术, 2022, 48(8): 3250-3259.
WU C H, DONG A L, LI Z H, et al.Photovoltaic power prediction based on graph similarity day and PSO-XGBoost[J]. High voltage engineering, 2022, 48(8): 3250-3259.
[22] 周丽婷. 风电叶片缺陷电磁超声检测技术研究[D]. 沈阳: 沈阳工业大学, 2022.
ZHOU L T.Research on electromagnetic ultrasonic detection technology of wind power blade defects[D]. Shenyang: Shenyang University of Technology, 2022.