提出了一种基于多重分形与近似熵相结合的信号特征量提取方法，应用于齿轮箱的故障信号诊断中。针对齿轮箱的故障信号的复杂性，先用减法聚类对提取到的信号特征量进行处理，得到初始聚类中心，然后再用模糊C 均值聚类(FCM)作进一步处理，实现齿轮箱故障的自动诊断和识别。多重分形谱提取的特征量如谱宽，可以表示信号的波动程度，而近似熵可以表示信号的复杂程度。两者结合可以得到更加准确的齿轮箱故障信号模式。减法聚类可以有效解决FCM容易陷入局部最优的问题，还可以提高收敛速度。提取的特征参数作为聚类分析的数据，通过计算数据点与聚类中心的隶属度判定所属类型，实现齿轮箱故障类型聚类以及模式识别。通过风力发电机齿轮箱故障诊断实验，证明了该方法的可行性和有效性。为齿轮箱故障诊断提供了一种新的有效途径。
Abstract
A feature extraction method based on multifractal and approximate entropy is presented, which is used for gearbox fault signal. Because of the complexity of gearbox fault data, firstly subtractive clustering is used to obtain the initial cluster center of characteristics. Then fuzzy c-means clustering(FCM) is used for further processing to achieve automatic gearbox fault diagnosis and identification．The volatility of a signal can be expressed by feature values extracted by multifractal spectrum, such as spectral width; and the complexity of the signal can be represented by approximate entropy(ApEn)．Combination of above obtains patterns of gearbox faults more accurately．Problem that FCM is easily fallen into local optimum can be effectively solved by subtractive fuzzy clustering, which also improves convergence rate. Characteristic parameters are processed by clustering analysis. In order to achieve gearbox fault clustering and recognition, the membership grade of data points and cluster center is calculated. To prove the feasibility and effectiveness of the method proposed in this paper, wind turbine gearbox fault diagnosis experiment is implemented. This paper provides a new effective way for gearbox fault diagnosis.
关键词
多重分形 /
近似熵 /
减法模糊聚类 /
故障诊断
{{custom_keyword}} /
Key words
mutifractal /
approximate entropy /
subtractive fuzzy clustering /
fault diagnosis
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 林近山, 陈前．基于非平稳时间序列双标度指数特征的齿轮箱故障诊断[J]．机械工程学报, 2012, 48(13): 108- 114．
LIN Jin-shan, CHEN Qian. Fault diagnosis of gearboxes based on the double-scaling-exponent characteristic of non-stationary time series [J]. Journal of Mechanical Engineering, 2012, 48(13): 108- 114．
[2] 唐新安, 谢志明, 王哲等． 风力机齿轮箱故障诊断[J]．噪声与振动控制, 2007(01): 120-124．
TANG Xin-an, XIE Zhi-ming, WANG Zhe, et al. Fault diagnosis of gearbox for wind turbine [J]. Noise and Vibration Control, 2007(01): 120-124．
[3] 祝志慧, 孙云莲， 李洪． 基于EMD的时频分析方法的电力故障信号检测[J]．武汉大学学报(工学版), 2007, 40(05): 112-119．
ZHU Zhi-hui, SUN Yun-lian, LI Hong. Power fault detection using empirical mode decomposition [J].Engineering Journal of Wuhan University, 2007, 40(05): 112-119．
[4] Frei M G，Osorio Ivan．Intrinsic time-scale decomposition: time-frequency-energy analysis and real-time filtering of non-stationary signals [J]．Proceedings of the Royal Society A, 2007, 463(2078): 321-342．
[5] 林近山，陈前．基于多重分形去趋势波动分析的齿轮箱故障特征提取方法[J]．振动与冲击, 2013, 32(02): 97-101．
LIN Jin-shan, CHEN Qian. Fault feature extraction of gearboxes based on multi-fractal detrended fluctuation analysis[J]. Journal of Vibration and Shock, 2013, 32(02): 97-101．
[6] 胥永刚, 李凌均, 何正嘉． 近似熵及其在机械设备故障诊断中的应用[J]．信息与控制, 2002, 31(06): 543-551．
XU Yong-gang, LI Ling-jun and HE Zheng-jia. Approximate entropy and its applications in mechanical fault diagnosis[J]. Information and Control, 2002, 31(06): 543-551．
[7] BEZDEK J C．Pattern recognition with fuzzy objective function slgorithms[M].New York：Plenum Press,1981．
[8] 王祖林,周荫清. 多重分形谱及其计算[J]. 北京航空航天大学学报, 2000,26(03): 256-258.
Wan Zu-lin, Zhou Yin-qing. Multifractal spectrum and calculation[J]. Journal of Beijing University of Aeronautics and Astronautics, 2000, 26(03): 256-258．
[9] 戴桂平．基于EMD近似熵和DAGSVM的机械故障诊断[J]．机械强度, 2010, 31(05): 467-471．
DAI Gui-Ping. Mechanical fault intelligent diagnosis based on EMD-Approximate entropy and LS-SVM[J]. Journal of Mechanical Strength, 2010, 31(05): 467-471．
[10] Abásolo D, Hornero R, Espino P, etal．Analysis of regularity in the EEG background activity of Alzheimer’s disease patients with Approximate Entropy[J]．Clinical Neurophysiology, 2005, (116): 1826-1834．
[11] 肖春景, 张敏. 基于减法聚类与模糊c-均值的模糊聚类的研究[J]. 计算机工程, 2005, 31(07): 135-137.
XIAO Chun-jing, ZHANG Min. Research on fuzzy clustering based on subtractive clustering and fuzzy c-means[J]. Computer Engineering, 2005, 31(07): 135-137.
[12] 蔡威, 程俊杰. 基于减法聚类的ＧＫ模糊聚类研究[J]. 兰州交通大学学报, 2011, 30(06): 50-54.
CAI Wei, CHENG Jun-jie. Research on GK fuzzy clustering algorithm based on subtractive clustering[J]. Journal of Lanzhou Jiaotong University, 2011, 30(06): 50-54.
[13] 徐超, 张培林, 任国全等． 基于改进半监督模糊C均值聚类的发动机磨损故障诊断[J]．机械工程学报, 2011, 47(17):55-60．
XU Chao, ZHANG Pei-lin, REN Guoquan, et al. Engine wear fault diagnosis based on improved semi-supervised fuzzy c-means clustering[J]. Journal of Mechanical Engineering, 2011, 47(17): 55-60．
[14] 张淑清, 张金敏, 赵玉春． 基于混沌和模糊聚类的机械故障自动识别[J]．机械工程学报, 2011, 47(19) :81-85．
ZHANG Shu-qing, ZHANG Jin-min, ZHAO Yu-chun, et al. Automatic diagnosis techniques of machinery fault based on chaos and fuzzy clustering analysis[J]. Journal of Mechanical Engineering, 2011, 47(19): 81-85
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1207 KB)
