基于改进复合多尺度样本熵的行星齿轮箱故障诊断

李伟; 王付广; 王东生

doi:10.13224/j.cnki.jasp.20220691

基于改进复合多尺度样本熵的行星齿轮箱故障诊断

doi: 10.13224/j.cnki.jasp.20220691

李伟^{1, 2,},
王付广¹,
王东生^{1, 2}

1.
铜陵学院机械工程学院，安徽铜陵 244000
2.
铜陵学院安徽铜基新材料产业共性技术研究中心，安徽铜陵 244000

基金项目: 国家自然科学基金（51205198）；安徽省自然科学基金（2008085ME149）；安徽省高校科研基金（2023AH051665）；平台建设协同创新项目（GXXT-2022-090）；校级自然科研项目（2022tlxy49）

详细信息

作者简介:
李伟（1993-），男，讲师，硕士，主要从事故障诊断和撞击动力学等方面的研究

中图分类号: V232.8；TH165⁺.3；TN911.7
计量
- 文章访问数: 61
- HTML浏览量: 78
- PDF量: 11
- 被引次数: 0
出版历程
- 收稿日期: 2022-09-15
- 网络出版日期: 2023-11-08

Fault diagnosis of planetary gearbox based on improved composite multi-scale sample entropy

LI Wei^{1, 2
,},
WANG Fuguang¹,
WANG Dongsheng^{1, 2}

1.
School of Mechanical Engineering，Tongling University，Tongling Anhui 244000，China
2.
Anhui Copper based New Materials Industry Common Technology Research Center，Tongling University，Tongling Anhui 244000，China

摘要

摘要:
针对多尺度样本熵受样本长度影响较大，且粗粒化过程较粗糙，易忽略有效信息的不足，在复合多尺度样本熵的基础上，以采样点间能量分布作为权重进行粗粒化计算，提出了改进的复合多尺度样本熵，并将其应用于行星齿轮箱故障诊断。通过仿真信号研究不同参数和不同噪声特性对改进复合多尺度样本熵算法的影响，将其与多尺度样本熵、广义多尺度样本熵、复合多尺度样本熵进行对比，验证了本文改进算法的稳定性。结合变分模态分解、主成分分析和支持向量机对行星齿轮箱实验信号进行故障诊断。对比结果表明：所提方法能够有效地实现不同工况和不同结构行星齿轮箱太阳轮常见故障诊断，且故障识别率达到95%以上，具有一定的有效性。
- 多尺度样本熵 /
- 变分模态分解 /
- 支持向量机 /
- 行星齿轮箱 /
- 故障诊断
Abstract:
In view of the fact that the multi-scale sample entropy is greatly affected by the sample length, the coarse graining process is relatively rough, and the shortage of effective information may be easily ignored, based on the composite multi-scale sample entropy, the energy distribution between sampling points was used as the weight for coarse graining calculation, and an improved composite multi-scale sample entropy was proposed and applied to the fault diagnosis of planetary gearbox. The influences of different parameters and noise characteristics on the improved composite multi-scale sample entropy algorithm were studied through simulation signals. The stability of the improved algorithm was verified by comparing it with multi-scale sample entropy, generalized multi-scale sample entropy and composite multi-scale sample entropy. Combined with variational mode decomposition, principal component analysis and support vector machine, the fault diagnosis of planetary gearbox experimental signals was carried out. The comparison results showed that the method can effectively realize the common fault diagnosis of the sun gear of the planetary gearbox under different working conditions and structures, and the fault identification rate was more than 95%, with certain effectiveness.
- multiscale sample entropy /
- variational modal decomposition /
- support vector machine /
- planetary gearbox /
- fault diagnosis

HTML

图 1 改进复合多尺度样本熵特征提取流程图

Figure 1. Flow chart of improved composite multi-scale sample entropy feature extraction

下载: 全尺寸图片幻灯片

图 2 改进粗粒化示意图

Figure 2. Schematic diagram of improved coarse granulation

下载: 全尺寸图片幻灯片

图 3 不同噪声背景下，样本长度N对MSE、GMSE、CMSE和ICMSE分析结果影响

Figure 3. Effect of sample length N on MSE, GMSE, CMSE and ICMSE analysis results under different noise backgrounds

下载: 全尺寸图片幻灯片

图 4 粉噪声背景下，不同相似容限r对MSE、GMSE、CMSE和ICMSE分析结果影响

Figure 4. Influence of different similarity tolerance r on MSE, GMSE, CMSE and ICMSE analysis results under thepowder noise background

下载: 全尺寸图片幻灯片

图 5 白噪声背景下，不同相似容限r对MSE、GMSE、CMSE和ICMSE分析结果影响

Figure 5. Influence of different similarity tolerance r on MSE, GMSE, CMSE and ICMSE analysis results under white noise background

下载: 全尺寸图片幻灯片

图 6 基于ICMSE的故障诊断流程图

Figure 6. ICMSE based fault diagnosis flow chart

下载: 全尺寸图片幻灯片

图 7 实验台示意图

Figure 7. Schematic diagram of test bench

下载: 全尺寸图片幻灯片

图 8 太阳轮故障示意图

Figure 8. Fault diagram of sun gear

下载: 全尺寸图片幻灯片

图 9 缺齿故障惩罚因子、分解维数与皮尔逊相关系数变化趋势

Figure 9. Change trend of penalty factor, decomposition dimension and Pearson correlation coefficient in normal state

下载: 全尺寸图片幻灯片

图 10 20 Hz转频下太阳轮常见故障不同熵值分析

Figure 10. Analysis of different entropy values of common faults of sun gear at 20 Hz rotating frequency

下载: 全尺寸图片幻灯片

图 11 30Hz转频下单级行星齿轮箱太阳轮常见故障ICMSE分析结果

Figure 11. ICMSE analysis results of faults of sun gear of single-stage planetary gearbox at 30 Hz rotating frequency

下载: 全尺寸图片幻灯片

图 12 20 Hz转频下双级行星齿轮箱太阳轮常见故障ICMSE分析结果

Figure 12. ICMSE analysis results of faults of sun gear of two-stage planetary gearbox at 20 Hz rotating frequency

下载: 全尺寸图片幻灯片

图 13 30 Hz转频下双级行星齿轮箱太阳轮常见故障ICMSE分析结果

Figure 13. ICMSE analysis results of faults of sun gear of two-stage planetary gearbox at 30 Hz rotating frequency

下载: 全尺寸图片幻灯片

图 14 MSE特征集降重后第一、二主成分分量

Figure 14. First and second principal components of MSE feature set after dimension reduction

下载: 全尺寸图片幻灯片

图 15 GMSE特征集降重后第一、二主成分分量

Figure 15. First and second principal components of GMSE feature set after dimension reduction

下载: 全尺寸图片幻灯片

图 16 CMSE特征集降重后第一、二主成分分量

Figure 16. First and second principal components of CMSE feature set after dimension reduction

下载: 全尺寸图片幻灯片

图 17 ICMSE特征集降重后第一、二主成分分量

Figure 17. First and second principal components of ICMSE feature set after dimension reduction

下载: 全尺寸图片幻灯片

表 1 20 Hz转频下太阳轮常见故障对应VMD参数

Table 1. VMD parameters corresponding to common faults of sun gear at 20 Hz rotating frequency

算法参数	正常	裂纹	缺齿	磨损
K	4	5	6	6
α	2250	2750	1750	1750

下载: 导出CSV

表 2 f_im与原始信号的相关系数和欧氏距离

Table 2. Correlation coefficient and Euclidean distance between f_im and original signal

分量	相关系数	欧氏距离
f_im1	0.8153	0.4892
f_im2	0.7703	0.6077
f_im3	0.7033	0.6924
f_im4	0.6567	0.7512
f_im5	0.5614	0.8092
f_im6	0.4756	0.8331

下载: 导出CSV

表 3 支持向量机多故障分类器对4种太阳轮故障识别结果（Q_train1，Q_test1）

Table 3. Recognition results of support vector machine multi fault classifier for four kinds of sun gear faults （Q_train1，Q_test1）

特征提取方法	（c，g）	SVM多故障分类器对测试样本识别结果/%
特征提取方法	（c，g）	正常状态识别率	裂纹故障识别率	缺齿故障识别率	磨损故障识别率	平均故障识别率
MSE	（48.50，0.57）	80	55	95	90	80
GMSE	（16，1.5）	90	90	100	55	83.75
CMSE	（27.8，7.5）	80	90	100	90	90
ICMSE	（62.7，3.2）	80	100	100	100	95

下载: 导出CSV

表 4 基于ICMSE和SVM的识别结果

Table 4. Recognition results based on ICMSE and SVM

特征提取方法	随机样本	（c，g）	SVM多故障分类器对测试样本识别结果/%
特征提取方法	随机样本	（c，g）	正常状态识别率	裂纹故障识别率	缺齿故障识别率	磨损故障识别率	平均故障识别率
ICMSE	Q_train2，Q_test2	（67，4.7）	87	100	100	97	96
ICMSE	Q_train3，Q_test3	（48，6.2）	83	100	100	100	95.75

下载: 导出CSV

参考文献(18)

[1]	周怡娜,董宏丽,张勇,等. 基于VMD去噪和散布熵的管道信号特征提取方法[J]. 吉林大学学报(工学版),2022,52(4): 959-969. ZHOU Yina,DONG Hongli,ZHANG Yong,et al. Feature extraction method of pipeline signals based on VMD denoising and dispersion entropy[J]. Journal of Jilin University (Engineering and Technology Edition),2022,52(4): 959-969. (in Chinese ZHOU Yina, DONG Hongli, ZHANG Yong, et al. Feature extraction method of pipeline signals based on VMD denoising and dispersion entropy[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(4): 959-969. (in Chinese)
[2]	党建,罗燚,田录林,等. 基于优化的VMD融合信息熵和FA_PNN的风电机组齿轮箱故障诊断[J]. 太阳能学报,2021,42(1): 198-204. DANG Jian,LUO Yi,TIAN Lulin,et al. Fault diagnosis of wind turbine gearbox based on optimized vmd fusion information entropy and FA_PNN[J]. Acta Energiae Solaris Sinica,2021,42(1): 198-204. (in Chinese DANG Jian, LUO Yi, TIAN Lulin, et al. Fault diagnosis of wind turbine gearbox based on optimized vmd fusion information entropy and FA_PNN[J]. Acta Energiae Solaris Sinica, 2021, 42(1): 198-204. (in Chinese)
[3]	雷亚国,何正嘉,林京,等. 行星齿轮箱故障诊断技术的研究进展[J]. 机械工程学报,2011,47(19): 59-67. LEI Yaguo,HE Zhengjia,LIN Jing,et al. Research advances of fault diagnosis technique for planetary gearboxes[J]. Journal of Mechanical Engineering,2011,47(19): 59-67. (in Chinese doi: 10.3901/JME.2011.19.059 LEI Yaguo, HE Zhengjia, LIN Jing, et al. Research advances of fault diagnosis technique for planetary gearboxes[J]. Journal of Mechanical Engineering, 2011, 47(19): 59-67. (in Chinese) doi: 10.3901/JME.2011.19.059
[4]	冯志鹏,褚福磊. 行星齿轮箱齿轮分布式故障振动频谱特征[J]. 中国电机工程学报,2013,33(2): 118-125,21. FENG Zhipeng,CHU Fulei. Vibration spectral characteristics of distributed gear fault of planetary gearboxes[J]. Proceedings of the CSEE,2013,33(2): 118-125,21. (in Chinese FENG Zhipeng, CHU Fulei. Vibration spectral characteristics of distributed gear fault of planetary gearboxes[J]. Proceedings of the CSEE, 2013, 33(2): 118-125, 21. (in Chinese)
[5]	HAN Minghong,PAN Jiali. A fault diagnosis method combined with LMD,sample entropy and energy ratio for roller bearings[J]. Measurement,2015,76: 7-19. doi: 10.1016/j.measurement.2015.08.019
[6]	WU Huachun,ZHOU Jian,XIE Chunhu,et al. Two-dimensional time series sample entropy algorithm: applications to rotor axis orbit feature identification[J]. Mechanical Systems and Signal Processing,2021,147: 107123. doi: 10.1016/j.ymssp.2020.107123
[7]	杨大为,冯辅周,赵永东,等. VMD样本熵特征提取方法及其在行星变速箱故障诊断中的应用[J]. 振动与冲击,2018,37(16): 198-205. YANG Dawei,FENG Fuzhou,ZHAO Yongdong,et al. A VMD sample entropy feature extraction method and its application in planetary gearbox fault diagnosis[J]. Journal of Vibration and Shock,2018,37(16): 198-205. (in Chinese YANG Dawei, FENG Fuzhou, ZHAO Yongdong, et al. A VMD sample entropy feature extraction method and its application in planetary gearbox fault diagnosis[J]. Journal of Vibration and Shock, 2018, 37(16): 198-205. (in Chinese)
[8]	雷春丽,曹鹏瑶,崔攀,等. 样本熵和VMD结合的轴承早期故障预测方法[J]. 机械设计与制造,2022(6): 300-304. LEI Chunli,CAO Pengyao,CUI Pan,et al. An early failure prediction method based on sample entropy and variational modal decomposition[J]. Machinery Design & Manufacture,2022(6): 300-304. (in Chinese LEI Chunli, CAO Pengyao, CUI Pan, et al. An early failure prediction method based on sample entropy and variational modal decomposition[J]. Machinery Design & Manufacture, 2022(6): 300-304. (in Chinese)
[9]	刘建昌,权贺,于霞,等. 基于参数优化VMD和样本熵的滚动轴承故障诊断[J]. 自动化学报,2022,48(3): 808-819. LIU Jianchang,QUAN He,YU Xia,et al. Rolling bearing fault diagnosis based on parameter optimization VMD and sample entropy[J]. Acta Automatica Sinica,2022,48(3): 808-819. (in Chinese LIU Jianchang, QUAN He, YU Xia, et al. Rolling bearing fault diagnosis based on parameter optimization VMD and sample entropy[J]. Acta Automatica Sinica, 2022, 48(3): 808-819. (in Chinese)
[10]	COSTA M,GOLDBERGER A L,PENG C K. Multiscale entropy analysis of complex physiologic time series[J]. Physical Review Letters,2002,89(6): 068102. doi: 10.1103/PhysRevLett.89.068102
[11]	郑近德,潘海洋,张俊,等. 基于多变量多尺度模糊熵的行星齿轮箱故障诊断[J]. 振动与冲击,2019,38(6): 187-193. ZHENG Jinde,PAN Haiyang,ZHANG Jun,et al. Multivariate multiscale fuzzy entropy based planetary gearbox fault diagnosis[J]. Journal of Vibration and Shock,2019,38(6): 187-193. (in Chinese ZHENG Jinde, PAN Haiyang, ZHANG Jun, et al. Multivariate multiscale fuzzy entropy based planetary gearbox fault diagnosis[J]. Journal of Vibration and Shock, 2019, 38(6): 187-193. (in Chinese)
[12]	董治麟,郑近德,潘海洋,等. 一种时移多尺度排列熵与ELM相结合的滚动轴承故障诊断方法[J]. 机械科学与技术,2021,40(10): 1523-1529. DONG Zhilin,ZHENG Jinde,PAN Haiyang,et al. A rolling bearing fault diagnosis method of time-shifted multi-scale permutation entropy combining with ELM[J]. Mechanical Science and Technology for Aerospace Engineering,2021,40(10): 1523-1529. (in Chinese DONG Zhilin, ZHENG Jinde, PAN Haiyang, et al. A rolling bearing fault diagnosis method of time-shifted multi-scale permutation entropy combining with ELM[J]. Mechanical Science and Technology for Aerospace Engineering, 2021, 40(10): 1523-1529. (in Chinese)
[13]	王振亚,姚立纲. 广义精细复合多尺度样本熵与流形学习相结合的滚动轴承故障诊断方法[J]. 中国机械工程,2020,31(20): 2463-2471. WANG Zhenya,YAO Ligang. Rolling bearing fault diagnosis method based on generalized refined composite multiscale sample entropy and manifold learning[J]. China Mechanical Engineering,2020,31(20): 2463-2471. (in Chinese WANG Zhenya, YAO Ligang. Rolling bearing fault diagnosis method based on generalized refined composite multiscale sample entropy and manifold learning[J]. China Mechanical Engineering, 2020, 31(20): 2463-2471. (in Chinese)
[14]	吴守军,陈健,冯辅周,等. 改进的熵特征在行星齿轮裂纹故障识别中的应用[J]. 西安交通大学学报,2021,55(6): 61-68. WU Shoujun,CHEN Jian,FENG Fuzhou,et al. An application of improved entropy feature in crack fault identification of planetary gear[J]. Journal of Xi’an Jiaotong University,2021,55(6): 61-68. (in Chinese WU Shoujun, CHEN Jian, FENG Fuzhou, et al. An application of improved entropy feature in crack fault identification of planetary gear[J]. Journal of Xi’an Jiaotong University, 2021, 55(6): 61-68. (in Chinese)
[15]	李海平,赵建民,张鑫,等. 行星齿轮箱齿轮磨损故障诊断[J]. 振动与冲击,2019,38(23): 84-89,125. LI Haiping,ZHAO Jianmin,ZHANG Xin,et al. Fault diagnosis for gear wear of planetary gearbox[J]. Journal of Vibration and Shock,2019,38(23): 84-89,125. (in Chinese LI Haiping, ZHAO Jianmin, ZHANG Xin, et al. Fault diagnosis for gear wear of planetary gearbox[J]. Journal of Vibration and Shock, 2019, 38(23): 84-89, 125. (in Chinese)
[16]	黄宇斐,石新发,贺石中,等. 一种基于主成分分析与支持向量机的风电齿轮箱故障诊断方法[J]. 热能动力工程,2022,37(10): 175-181. HUANG Yufei,SHI Xinfa,HE Shizhong,et al. A fault diagnosis method of wind turbine gearbox based on PCA and SVM[J]. Journal of Engineering for Thermal Energy and Power,2022,37(10): 175-181. (in Chinese HUANG Yufei, SHI Xinfa, HE Shizhong, et al. A fault diagnosis method of wind turbine gearbox based on PCA and SVM[J]. Journal of Engineering for Thermal Energy and Power, 2022, 37(10): 175-181. (in Chinese)
[17]	郑近德,刘涛,孟瑞,等. 基于广义复合多尺度排列熵与PCA的滚动轴承故障诊断方法[J]. 振动与冲击,2018,37(20): 61-66. ZHENG Jinde,LIU Tao,MENG Rui,et al. Generalized composite multiscale permutation entropy and PCA based fault diagnosis of rolling bearings[J]. Journal of Vibration and Shock,2018,37(20): 61-66. (in Chinese ZHENG Jinde, LIU Tao, MENG Rui, et al. Generalized composite multiscale permutation entropy and PCA based fault diagnosis of rolling bearings[J]. Journal of Vibration and Shock, 2018, 37(20): 61-66. (in Chinese)
[18]	但长林,李三雁,张彬. 基于样本熵和SVM的滚动轴承故障诊断方法研究[J]. 中国测试,2020,46(11): 37-42. DAN Changlin,LI Sanyan,ZHANG Bin. Research on fault diagnosis method for rolling element bearings based on sample entropy and SVM[J]. China Measurement & Test,2020,46(11): 37-42. (in Chinese DAN Changlin, LI Sanyan, ZHANG Bin. Research on fault diagnosis method for rolling element bearings based on sample entropy and SVM[J]. China Measurement & Test, 2020, 46(11): 37-42. (in Chinese)