1. College of Mechanical and Power Engineering, Nanjing Tech University, Nanjing 211816, China 2. Jiangsu Key Laboratory of Digital Manufacturing for Industrial Equipment and Control Technology, Nanjing 211816, China 3. Minth Group, Ningbo 315806, China
Multi-fractal adaptive feature extraction method based on Wavelet leader method and isometric mapping method optimized by hybrid grey wolf optimization algorithm (HGWO-ISOMAP) was proposed, in order to solve the problem that the vibration signal of slewing bearing is weak and the feature information is difficult to extract. Wavelet leader is utilized to calculate the multi-fractal features, mine the geometric structure information of vibration data, and construct a high-dimensional multi-fractal feature matrix. Adaptive feature selection of high-dimensional feature matrix is carried out through ISOMAP method optimized by HGWO. The selected feature matrix is input into the least squares support vector machine (LSSVM) optimized by genetic algorithm (GA) for fault state identification. A full life experiment of a certain type of slewing bearing was carried out by using self-developed comprehensive performance test platform of slewing bearing, in order to verify the superiority of the proposed method. Results show that compared with general time domain, time-frequency domain and frequency domain feature extraction methods, the proposed method can improve the recognition accuracy and reduce the calculation time, providing a new effective way for feature extraction of slewing bearing.
Fig.1Flow chart of adaptive feature extraction method of slewing bearing
Fig.2Physical map of slewing bearing test rig
Fig.3System structure flow chart of slewing bearing test rig
参数
数值
参数
数值
滚道中心直径/mm
1 000
滚珠数目
71
齿数
96
螺栓个数
36
滚珠直径/mm
40
轴/径向间隙/mm
0.05~0.20
外圈外径/mm
1 185.6
内圈内径/mm
878
钢球材料
GCr15
内/外圈材料
42GMo
Tab.1Structural parameters of a certain type of slewing bearing
Fig.4Acceleration signal map of slewing bearing
Fig.5Comparison of vibration signal before and after noise reduction
Fig.6Multifractal feature map in three states
Fig.7Multifractal spectrum-singular exponential graph of a single sample
Fig.8Statistical results of special point distribution in three states
状态
初始点范围
最高点范围
终止点范围
正常状态
0.05~0.15
0.22~0.29
0.32~0.40
螺栓破坏
0.04~0.07
0.07~0.10
0.12~0.20
外圈破坏
0.35~0.54
0.57~0.65
0.72~0.88
Tab.2Statistical results of special point distribution intervals
Fig.9Dimension reduction results of HGWO-ISOMAP method
Fig.10Data visualization after dimensionality reduction
Fig.11Optional three-dimensional data visualization after dimensionality reduction
Fig.12Classification results of LS-SVM method
分类器
方法
Rc/%
t/s
LSSVM
HGWO-ISOMAP-Wavelet leader
99.33
361.453
降维前-Wavelet leader
96.67
560.643
f1-f2-f3
98.00
374.419
f1-f2-f4
98.67
360.888
f2-f3-f4
88.00
366.519
Tab.3Recognition results of proposed method
分类器
方法
Rc/%
t /s
LSSVM
时域特征
85.33
377.801
频域特征
88.00
366.634
时频域特征
89.33
352.214
时域-频域-时频域混合特征
92.00
417.547
BP神经网络
HGWO-ISOMAP-Wavelet leader
88.67
30.158
降维前-Wavelet leader
82.67
217.013
f1-f2-f3
86.00
28.765
f1-f2-f4
88.45
29.127
f2-f3-f4
67.33
29.333
时域特征
66.00
17.497
频域特征
66.67
21.377
时频域特征
68.67
20.046
时域-频域-时频域混合特征
75.33
23.223
Tab.4Recognition results of other methods
[1]
张慧芳, 陈捷 大型回转支承故障信号处理方法综述[J]. 机械设计与制造, 2012, (3): 216- 218 ZHANG Hui-fang, CHEN Jie Research on signal processing method of large slewing bearing[J]. Machinery Design and Manufacture, 2012, (3): 216- 218
doi: 10.3969/j.issn.1001-3997.2012.03.080
[2]
刘志军.风电回转支承监测与故障诊断研究[D]. 南京: 南京工业大学, 2011. LIU Zhi-jun. Research of monitoring and fault diagnosis of slewing bearing in wind turbine [D]. Nanjing: Nanjing Tech University, 2011.
[3]
JAOUHER B, NADER F, LOTFI S, et al Application of empirical mode decomposition and artificial neural network for automatic bearing fault diagnosis based on vibration signals[J]. Applied Acoustics, 2015, 89 (3): 16- 27
[4]
YANG Y, YU D, CHENG J A roller bearing fault diagnosis method based on EMD energy entropy and ANN[J]. Journal of Sound and Vibration, 2006, 294 (1/2): 269- 277
[5]
魏永合, 王明华 基于EEMD和SVM的滚动轴承退化状态识别[J]. 计算机集成制造系统, 2015, 21 (9): 2475- 2483 WEI Yong-he, WANG Ming-hua Degradation state recognition of rolling bearing based on EEMD and SVM[J]. Computer Integrated Manufacturing Systems, 2015, 21 (9): 2475- 2483
[6]
陆超, 陈捷, 洪荣晶 采用概率主成分分析的回转支承寿命状态识别[J]. 西安交通大学学报, 2015, 49 (10): 90- 96 LU Chao, CHEN Jie, HONG Rong-jing Recognition of life state for slewing bearings using probabilistic component analysis[J]. Journal of Xi’an Jiaotong University, 2015, 49 (10): 90- 96
doi: 10.7652/xjtuxb201510015
[7]
LU C, CHEN J, HONG R, et al Degradation trend estimation of slewing bearing based on LSSVM model[J]. Mechanical Systems and Signal Processing, 2016, 76/77: 353- 366
doi: 10.1016/j.ymssp.2016.02.031
[8]
张淑清, 李盼, 胡永涛, 等 多重分形近似熵与减法FCM聚类的研究及应用[J]. 振动与冲击, 2015, 34 (18): 205- 209 ZHANG Shu-qing, LI Pan, HU Yong-tao, et al Application of multifractal approximate entropy and subtractive FCM clustering in gearbox fault diagnosis[J]. Journal of Vibration and Shock, 2015, 34 (18): 205- 209
[9]
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 (2): 97- 101 LIN Jin-shan, CHEN Qian Fault feature extraction of gearboxes based on multifractal detrended fluctuation analysis[J]. Journal of Vibration and Shock, 2013, 32 (2): 97- 101
doi: 10.3969/j.issn.1000-3835.2013.02.019
[10]
LIN J, CHEN Q Fault diagnosis of rolling bearings based on multifractal detrended fluctuation analysis and Mahalanobis distance criterion[J]. Mechanical Systems and Signal Processing, 2013, 38 (2): 515- 533
doi: 10.1016/j.ymssp.2012.12.014
[11]
XIONG Q, ZHANG W, LU T, et al A fault diagnosis method for rolling bearings based on feature fusion of multifractal detrended fluctuation analysis and alpha stable distribution[J]. Shock and Vibration, 2016, (3): 1- 12
[12]
WENDT H, ABRY P. Bootstrap for multifractal analysis [C] // 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings. Toulous: Institute of Electrical and Electronics Engineers, 2006: 38-48.
[13]
WENDT H, ABRY P. Bootstrap tests for the time constancy of multifractal attributes [C]// 2008 IEEE International Conference on Acoustics, Speech and Signal Processing. Las Vegas: IEEE, 2008: 3465-3468.
[14]
LASHERMES B, JAFFARD S, ABRY P. Wavelet leader based multifractal analysis [C]// 2005 IEEE International Conference on Acoustics, Speech, and Signal Processing. Philadelphia: IEEE, 2005: 161-164.
[15]
陈法法, 汤宝平, 苏祖强 基于等距映射与加权KNN的旋转机械故障诊断[J]. 仪器仪表学报, 2013, 34 (1): 215- 220 CHEN Fa-fa, TANG Bao-ping, SU Zu-qiang Rotating machinery fault diagnosis based on isometric mapping and weighted KNN[J]. Journal of Instrument and Meter, 2013, 34 (1): 215- 220
doi: 10.3969/j.issn.0254-3087.2013.01.031
[16]
计会凤, 徐爱功, 隋达嵬 Dijkstra算法的设计与实现[J]. 辽宁工程技术大学学报: 自然科学版, 2008, 27 (增1): 222- 223 JI Hui-feng, XU Ai-gong, SUI Da-wei Design and implementation of Dijkstra algorithm[J]. Journal of Liaoning Technical University: Natural Science, 2008, 27 (增1): 222- 223
[17]
周志华.机器学习[M]. 北京: 清华大学出版社, 2016: 226-231.
[18]
惠康华, 肖柏华, 王春恒 基于自适应近邻参数的局部线性嵌入[J]. 模式识别与人工智能, 2010, 23 (6): 842- 846 HUI Kang-hua, XIAO Bai-hua, WANG Chun-heng Self-regulation of neighborhood parameter for locally linear embedding[J]. Pattern Recognition and Artificial Intelligence, 2010, 23 (6): 842- 846
doi: 10.3969/j.issn.1003-6059.2010.06.015
[19]
MIRJALILI S, MIRJALILI S M, LEWIS A Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69 (3): 46- 61
[20]
高岳林, 刘俊梅 一种带有随机变异的动态差分进化算法[J]. 计算机应用, 2009, 29 (10): 2719- 2722 GAO Yue-lin, LIU Jun-mei Dynamic differential evolution algorithm with random mutation[J]. Journal of Computer Applications, 2009, 29 (10): 2719- 2722
[21]
陈仁祥, 汤宝平, 吕中亮 基于相关系数的EEMD转子振动信号降噪方法[J]. 振动、测试与诊断, 2012, 32 (4): 542- 546 CHEN Ren-xiang, TANG Bao-ping, LV Zhong-liang Ensemble empirical mode decomposition de-noising method based on correlation coefficients for vibration signal of rotor system[J]. Journal of Vibration, Measurement and Diagnosis, 2012, 32 (4): 542- 546
doi: 10.3969/j.issn.1004-6801.2012.04.004
[22]
王志华, 张建峰 基于EEMD降噪与PNN的齿轮箱齿轮故障诊断[J]. 煤矿机械, 2015, 36 (11): 326- 328 WANG Zhi-hua, ZHANG Jian-feng Fault diagnosis of gearbox gear based on EEMD de-noising and PNN[J]. Coal Mine Machinery, 2015, 36 (11): 326- 328
[23]
时培明, 梁凯, 赵娜, 等 基于深度学习特征提取和粒子群支持向量机状态识别的齿轮智能故障诊断[J]. 中国机械工程, 2017, 28 (9): 1056- 1061 SHI Pei-ming, LIANG Kai, ZHAO Na, et al Intelligent fault diagnosis for gears based on deep learning feature extraction and particle swarm optimization SVM state identification[J]. China Mechanical Engineering, 2017, 28 (9): 1056- 1061
doi: 10.3969/j.issn.1004-132X.2017.09.009
[24]
张梅军, 王闯, 陈灏 IMF能量和RBF神经网络相结合在滚动轴承故障诊断中的应用研究[J]. 机械, 2012, 39 (6): 63- 66 ZHANG Mei-jun, WANG Chuang, CHEN Hao The application research on the combination of IMF energy and RBF neural network in rolling bearing fault diagnosis[J]. Mechanics, 2012, 39 (6): 63- 66