针对风电机组齿轮箱结构复杂、受交变载荷和恶劣工作环境影响容易出现故障导致停机的问题,提出基于统计学K-均值聚类理论的统计型监督式局部线性嵌入流形学习（S-SLLE）特征维数约简方法,首先通过对齿轮箱振动信号时频域故障特征提取,剔除冗余特征向量,减少诊断模型的复杂度和计算量,再利用RBF核支持向量机分类器建立诊断模型,对S-SLLE提取的特征向量进行分类识别,以提高故障诊断模型的识别率。最后利用MFS机械故障模拟综合实验系统进行齿轮箱多类振动故障实验,通过对其实验故障信号的分析处理,其诊断实例结果验证了提出的S-SLLE RBF-SVM诊断模型能准确有效地进行风电机组齿轮箱故障诊断识别。
Abstract
Because of the complicated structure of wind turbine gearbox, it is easy to be shut down due to the influence of alternating load and harsh working environment. In order to improve the recognition rate of fault diagnosis model, the feature dimension reduction method of the statistical supervised locally linear embedding manifold learning（S-SLLE） based on K-means classification theory was proposed. Firstly, the time-frequency domain fault features of gearbox vibration signals are extracted, and the redundancy feature vector are taken out, so the complexity and calculation amount of the diagnosis model are reduced,then the diagnosis model based on the RBF kernel support vector machine classifier is used to establish to diagnose and identify the feature vector extracted by S-SLLE. Finally, the Machinery Fault Simulator was used to simulate multiple vibration fault experiments on the gearbox. Through the analysis and processing of the experimental fault signals, the results verify that the proposed S-SLLE RBF-SVM diagnosis model can identify the wind turbine gearbox fault effectively and accurately.
关键词
风电机组 /
特征提取 /
支持向量机 /
流形学习 /
齿轮箱振动故障
Key words
wind turbines /
feature extraction /
support vector machines /
manifold learning /
gearbox vibration fault
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 陈雪峰, 李继猛, 程航, 等. 风力发电机状态监测和故障诊断技术的研究与进展[J]. 机械工程学报, 2011, 47(9): 45-52.
CHEN X F, LI J M, CHENG H, et al.Research and application of condition monitoring and fault diagnosis technology in wind turbines[J]. Journal of mechanical engineering, 2011, 47(9): 45-52.
[2] RIBRANT J, BERTLING L M.Survey of failures in wind power systems with focus on Swedish wind power plants during 1997-2005[J]. IEEE transactions on energy conversion, 2007, 22(1): 167-173.
[3] MALHI A, GAO R X.PCA-based feature selection scheme for machine defect classification[J]. IEEE transactions on instrumentation and measurement, 2004, 53(6): 1517-1525.
[4] JOLLIFFE I T.Principal component analysis[M]. New York: Springer, 1986.
[5] WEBB A R.Introduction to statistical pattern recognition[M]. Second Edition. New York: Academic Press, 2003.
[6] COX T F, COX M A A. Multi-dimensional scaling[M]. London: Chapman & Hall, 1994.
[7] HYVARINEN A, OJA E.Independent component analysis: algorithms and applications[J]. Neural networks, 2000, 425(13): 411-430.
[8] CUETO E, CHINESTA F.Meshless methods for the simulation of material forming[J]. International journal of material forming, 2015, 8(1): 25-43.
[9] ZHANG K, KWOK J T.Clustered nyström nethod for large Scale manifold learning and dimension reduction[J]. IEEE transactions on neural networks, 2010, 21(10): 1576-1587.
[10] RADUCANU B, DORNAIKA F.A supervised non-linear dimensionality reduction approach for manifold learning[J]. Pattern recognition, 2012, 45(6): 2432-2444.
[11] GENG X, ZHAN D C, ZHOU Z H.Supervised nonlinear dimensionality reduction for visualization and classification[J]. IEEE transactions on systems man & cybernetics, Part B: Cybernetics, 2005, 35(6): 1098-1107.
[12] YANG X, FU H Y, ZHA H Y, et al.Semi-supervised nonlinear dimensionality reducation[C]//Proceeding of the 23th International Conference on Machine Learning, 2006: 1065-1072.
[13] CHEN H T, CHANG H W, LIU T L.Local discriminant embedding and its variants[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR’05), San Diego, USA, 2005: 846-853.
[14] YAN S C, XU D, ZHANG B Y, et al.Graph embedding and extensions: a general framework for dimensionality reduction[J]. IEEE transactions on pattern analysis and machine intelligence, 2007, 29(1): 40-51.
[15] 王燕, 苏文君, 刘花丽. 基于监督判别局部保持投影的表情识别算法[J]. 计算机工程与应用, 2014, 50(1): 195-199.
WANG Y, SU W J, LIU H L.Expression recognition algorithm based on supervised discriminat locality preserving projection[J]. Computer engineering and applications, 2014, 50(1): 195-199.
[16] ROWEIS S T, SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000, 290(5500): 2323-2326.
[17] KOUROPTEVA O, OKUN O, PIETIKAINEN M.Supervised locally linear embedding algorithm for pattern recognition[J]. Pattern recognition and image analysis, 2003, 2652(9): 386-394.
[18] XIAO J Z, XIAO L.A research of the partition clustering algorithm[C]//International Symposium Intelligence Information Processing and Trusted Computing, Huanggang, China, 2010: 551-553.
[19] SPECHT D F.Generation of polynomial discriminant functions for pattern recognition[J]. IEEE transactions on electronic computers, 1966, 16(3): 308-319.
[20] LI B, CHOW M Y, TIPSUWAN Y, et al.Neural-network-based motor rolling bearing fault diagnosis[J]. IEEE transactions on industrial electronics, 2000, 47(5): 1060-1069.
[21] BURGES C J C. A tutorial on support vector machines for pattern recognition[J]. Data mining & knowledge discovery, 1998, 2(2): 121-167.
[22] HSU C W, CHANG C C, LIN C J.A practical guide to support vector machines[J]. Machine learning, 2003, 180(2): 1-28.
基金
南京工程学院科研基金（ZKJ201606; ZKJ201703）
PDF(2033 KB)
