计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240900138-8.doi: 10.11896/jsjkx.240900138
王瑞, 汤占军
WANG Rui, TANG Zhanjun
摘要： 针对无人机在风机叶片表面缺陷检测中遇到的复杂特征处理和多形式缺陷表现不佳的问题,提出了一种基于多特征融合与集成学习的风机叶片缺陷检测方法。该方法通过提取局部LBP特征、HOG特征以及胶囊网络的高级特征,并将其进行有效融合,构建了一个多特征提取模型,以获取更深入的细节信息。同时,选择了3种具有不同偏差和方差特性的基础分类器——支持向量机(SVM)、k近邻算法(KNN)和决策树(DT),通过整合不同基模型的优势,建立异质集成学习模型,从而提升了模型的整体性能。在风机叶片表面缺陷图像数据集上对模型(MFEM)进行了验证,实验结果表明,该方法的平均精确度(MAP)最高达到98%,相比于YOLOv7和Faster R-CNN分别提高了3.1%和5.8%,对比SVM,KNN和DT 3类基模型有较大提升。此外,通过消融实验对不同模块的有效性进行了验证。实验结果表明,提出的多特征融合与集成学习模型(MFEM)在风机叶片缺陷检测任务中表现出了优良的性能。
中图分类号:
| [1]LEE J,ZHAO F.Global Wind Report 2022,Global Wind Energy Council(GWEC)[R].2018:5-6. [2]ZHAO X,MA X,CHENB,et al.Challenges toward carbon neutrality in China:Strategies and countermeasures[J].Resources,Conservation and Recycling,2022,176:105959. [3]RIBRANT J,BERTLING L.Survey of failures in wind power systems with focus on Swedish wind power plants during 1997－2005[C]//Proceedings of the 2007 I-EEE Power Engineering Society General Meeting.Tampa,FL,USA,2007:1-8 [4]MISHNAEVSKY J L.Sustainable end-of-life manage-ment ofwind turbine blades:Overview of current and coming solutions[J].Materials,2021,14(5):1124. [5]DING S;YANG C,ZHANG S.Acoustic-Signal-Based Damage Detection of Wind Turbine Blades－A Review[J].Sensors,2023,23:4987. [6]WANG W,XUE Y,HE C,et al.Review of the Typical Damage and Damage-Detection Methods of Large Wind Turbine Blades[J].Energies,2022,15:5672. [7]KHAZAEE M,DERIAN P,MOURAUD A.A comprehensivestudy on Structural Health Monitoring(SHM) of wind turbine blades by instrumenting tower using machine learning methods[J].Renew.Energy,2022,199:1568-1579. [8]TAN X G,ZHANG G M.The fan blade surface defect detection technology based on unmanned aerial vehicle(uav) inspection [J].Electric Measurement and Instrumentation,2025,62(3):183-189. [9]CAO Q C,WU L D,ZHANG L N,et al.Fan blades defect detection based on machine vision [J].Electric Technology,2021(22):74-76,155. [10]FAN H Y,HU X L,MA Y Q.Fan blade defect detection based on improved YOLOv7 and UAV aerial photography technology [J].The modeling and simulation,2023,12(5):4855-4867. [11]DAVIS M,NAZARIO DEJESUS E,SHEKARAMIZ M,et al.Identification and Localization of Wind Turbine Blade Faults Using Deep Learning[J].Appl.Sci.,2024,14:6319. [12]ZOU L,CHENG H,SUN Q.Surface Damage Identification of Wind Turbine Blade Based on Improved Lightweight Asymmetric Convolutional Neural Network[J].Appl.Sci.,2023,13:6330. [13]SHIHAVUDDIN A,CHEN X,FEDOROV V,et al.Wind Tu-rbine Surface Damage Detection by Deep Learning Aided Drone Inspection Analysis[J].Energies,2019,12:676. [14]ZHAO J,CHEN B,LI Y Z,et al.Acoustical Crack Feature Extraction of Tur-bine Blades under Complex Background Noise[J].Journal of Beijing University of Posts and Telecom,2017,40(5):117-122. [15]YANG J C,HAN S J,MAO L,et al.Review ofcapsule network[J].Journal of Shandong University(Engineering Science),2019,49(6):1-10. [16]AFSHAR P,HEIDARIAN S,NADERKHANI F,et al.Covid-caps:A capsule network-based framework for identification of covid-19 cases from xray images[J].Pattern Recognition Letters,2020,138:638-643. [17]LIAO J,LIN Y,MA T,et al.Facial expression recognitionmethods in the wild based on fusion feature of attention mechanism and LBP[J].Sensors,2023,23(9):4204. [18]SHARMA A K,NANDAL A,DHAKA A,et al.HOG transformation based feature extraction framework in modified Resnet50 model for brain tumor detection[J].Biomedical Signal Proces-sing and Control,2023,84:104737. [19]CHAABANE S B,HIJJI M,HARRABI R,et al.Face recogni-tion based on statistical features and SVM classifier[J].Multimedia Tools and Applications,2022,81(6):8767-8784. [20]GUO G,WANG H,BELL D,et al.KNN model-based approach in classification[C]//The Move to Meaningful Internet Systems 2003.CoopIS,DOA,and ODBASE:OTM Confederated International Conferences,CoopIS,DOA,and ODBASE 2003,Catania,Sicily,Italy,Springer Berlin Heidelberg,2003:986-996. [21]SUTHAHARAN S,SUTHAHARAN S.Decision tree learning[J].Machine Learning Models and Algorithms for Big Data Classification:Thinking with Examples for Effective Learning,2016,50(2):237-269. [22]SUYAL M,GOYAL P.A review on analysis of K-nearestneighbor classification machine learning algorithms based on supervised learning[J].International Journal of Engineering Trends and Technology,2022,70(7):43-48. [23]WEI J,CHU X,SUN X Y,et al.Machine learning in materials science[J].InfoMat,2019,1(3):338-358. [24]TRIPOLITI E E,FOTIADIS D I,MANIS G.Modifications ofthe construction and voting mechanisms of the random forests algorithm[J].Data & Knowledge Engineering,2013,87:41-65. |
|
||
首页