基于改进SSA-VMD和自相关滤波的滚动轴承故障诊断方法

doi:10.12305/j.issn.1001-506X.2026.09.20

摘要/Abstract

摘要：

针对滚动轴承故障诊断易受噪声干扰，同时变分模态分解（variationalmode decomposition, VMD）参数寻优易陷入局部最优且固有模态函数(intrinsic mode function, IMF)筛选依赖经验阈值，制约了故障特征的准确提取。为此，提出一种改进麻雀搜索算法（sparrowsearch algorithm, SSA）-VMD与自相关自适应滤波融合的滚动轴承故障诊断方法构建融合最小包络熵、峭度和包络谱峰值的适应度函数，并从种群初始化、发现者更新和追随者更新三个环节改进SSA，实现VMD模态数和惩罚因子的自适应寻优。随后，利用归一化自相关权重对各IMF进行自适应滤波、重构及包络谱分析，避免人工设置IMF筛选阈值。在3组公开轴承故障数据集上的实验表明，改进SSA相较于对应对比算法的收敛速度分别提高80%、20%和57%，同时获得更低的适应度值，所提方法的平均故障诊断准确率达到99.074%以上。结果表明，该方法能够降低VMD参数寻优和IMF筛选对人工经验的依赖，提高噪声条件下滚动轴承故障特征提取与诊断的准确性。

关键词: 改进麻雀搜寻算法, 参数寻优, 自适应滤波器设计, 信号处理, 故障诊断

Abstract:

Rolling bearing fault diagnosis is susceptible to noise interference, while variational mode decomposition（VMD） parameter optimization is prone to local optima and intrinsic mode function（IMF） selection relies heavily on empirical thresholds, thereby limiting the accurate extraction of fault features. To address these issues, an improved sparrow search algorithm（SSA）-VMD and autocorrelation adaptive-filtering method is proposed. A composite fitness function integrating minimum envelope entropy, kurtosis, and the crest factor of the envelope spectrum is constructed, and the Sparrow Search Algorithm is improved through population initialization as well as discoverer and follower updating strategies to adaptively optimize the VMD mode number and penalty factor. Subsequently, normalized autocorrelation weights are used to adaptively filter and reconstruct the IMFs, followed by envelope-spectrum analysis, thereby avoiding manual IMF-selection thresholds. Experiments on three public bearing-fault datasets show that the improved SSA increases convergence speed by 80%, 20%, and 57%, respectively, while achieving lower fitness values, and the proposed method obtains an average fault-diagnosis accuracy above 99.074%. These results demonstrate that the method reduces reliance on expert experience in VMD parameter optimization and IMF selection, while improving fault-feature extraction and diagnostic accuracy for rolling bearings under noisy conditions.

Key words: improved sparrow search algorithm (SSA), parameter optimization, adaptive filter design, signal processing, fault diagnosis

中图分类号:

桑硕海, 王浩伟, 赵建忠, 韩丹阳. 基于改进SSA-VMD和自相关滤波的滚动轴承故障诊断方法[J]. 系统工程与电子技术, 2026, 48(9): 3066-3078.

Shuohai Sang, Haowei Wang, Jianzhong Zhao, Danyang Han. Fault diagnosis method based on improved SSA-VMD and autocorrelation filtering for rolling bearing[J]. Systems Engineering and Electronics, 2026, 48(9): 3066-3078.

图/表 28

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

表1

图14

图16

图15

图17

图18

图19

图20

图21

表2

表3

表4

图22

图23

表5

参考文献 50

1	Gu Y K, Zhou X Q, Yu D P. Fault diagnosis method of rolling bearing using principal component analysis and support vector machine[J]. Journal of Mechanical Science and Technology, 2018, 32, 5079. doi: 10.1007/s12206-018-1004-0
2	Rai A, Upadhyay S H. A review on signal processing techniques utilized in the fault diagnosis of rolling element bearings[J]. Tribology International, 2016, 96, 289. doi: 10.1016/j.triboint.2015.12.037
3	Chen J J, Xu B, Zhang X. A vibration feature extraction method based on time-domain dimensional parameters and Mahalanobis distance[J]. Mathematical Problems in Engineering, 2021, 2021 (1): 2498178. doi: 10.1155/2021/2498178
4	Sun R, Yang Z, Chen X. Gear fault diagnosis based on the structured sparsity time-frequency analysis[J]. Mechanical Systems and Signal Processing, 2018, 102, 346. doi: 10.1016/j.ymssp.2017.09.028
5	Chen Z H, Cen J, Xiong J B. Rolling bearing fault diagnosis using time-frequency analysis and deep transfer convolutional neural network[J]. IEEE Access, 2020, 8, 150248. doi: 10.1109/ACCESS.2020.3016888
6	He M, He D. Deep learning based approach for bearing fault diagnosis[J]. IEEE Trans. on Industry Applications, 2017, 53 (3): 3057. doi: 10.1109/TIA.2017.2661250
7	刘长良, 武英杰, 甄成刚. 基于变分模态分解和模糊C均值聚类的滚动轴承故障诊断[J]. 中国电机工程学报, 2015, 35 (13): 3358. doi: 10.13334/j.0258-8013.pcsee.2015.13.020
8	赵昕海, 张术臣, 李志深, 等. 基于VMD的故障特征信号提取方法[J]. 振动、测试与诊断, 2018, 38 (1): 11. doi: 10.16450/j.cnki.issn.1004-6801.2018.01.002
9	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.
10	唐贵基, 王晓龙. 变分模态分解方法及其在滚动轴承早期故障诊断中的应用[J]. 振动工程学报, 2016, 29 (4): 638.
11	李波, 胡哿郗, 石剑钧, 等. 基于多惩罚因子优化VMD的滚动轴承故障特征提取方法[J]. 系统工程与电子技术, 2023, 45 (11): 3690.
12	褚惟, 王贵勇, 刘韬, 等. 麻雀算法参数优化VMD联合K-SVD的滚动轴承故障诊断[J]. 噪声与振动控制, 2022, 42 (4): 100.
13	刘成浩, 张晓林, 孙溶辰, 等. 基于改进粒子群的密度聚类算法混合矩阵估计[J]. 系统工程与电子技术, 2024, 46 (7): 2211. doi: 10.12305/j.issn.1001-506X.2024.07.05
14	欧阳鑫, 李志斌. 基于样本扩充和特征优选的IGWO优化SVM的变压器故障诊断技术[J]. 电力系统保护与控制, 2023, 51 (18): 11.
15	吕鑫, 慕晓冬, 张钧. 基于改进麻雀搜索算法的多阈值图像分割[J]. 系统工程与电子技术, 2021, 43 (2): 318. doi: 10.12305/j.issn.1001-506X.2021.02.05
16	刘玉明, 刘自然, 王鹏博. 基于ISSA-VMD的滚动轴承早期故障诊断方法[J]. 机电工程, 2023, 40 (9): 1426. doi: 10.3969/j.issn.1001-4551.2023.09.015
17	马增强, 李亚超, 刘政, 等. 基于变分模态分解和Teager能量算子的滚动轴承故障特征提取[J]. 振动与冲击, 2016, 35 (13): 134. doi: 10.13465/j.cnki.jvs.2016.13.022
18	张萍, 张文海, 赵新贺, 等. WOA-VMD算法在轴承故障诊断中的应用[J]. 噪声与振动控制, 2021, 41 (4): 86.
19	石文杰, 黄鑫, 温广瑞, 等. 基于DS-VMD及相关峭度的滚动轴承故障诊断[J]. 振动、测试与诊断, 2021, 41 (1): 133. doi: 10.16450/j.cnki.issn.1004-6801.2021.01.019
20	王琇峰, 文俊. 基于噪声信号和改进VMD的滚动轴承故障诊断[J]. 噪声与振动控制, 2021, 41 (2): 118. doi: 10.3969/j.issn.1006-1355.2021.02.019
21	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. doi: 10.16383/j.aas.c190345
22	王斐, 房立清, 赵玉龙, 等. 基于VMD和SVDD的滚动轴承早期微弱故障检测和性能退化评估研究[J]. 振动与冲击, 2019, 38 (22): 224.
23	万书亭, 豆龙江, 李聪, 等. 基于VMD和样本熵的高压断路器故障特征提取及分类[J]. 振动与冲击, 2018, 37 (20): 32. doi: 10.13465/j.cnki.jvs.2018.20.005
24	李宏, 李定文, 朱海琦, 等. 一种优化的VMD算法及其在语音信号去噪中的应用[J]. 吉林大学学报（理学版）, 2021, 59 (5): 1219. doi: 10.13413/j.cnki.jdxblxb.2020181
25	郑祥豪, 张宇宁, 李金伟, 等. 基于变分模态分解和互相关分析的旋转机械信号降噪研究[J]. 核科学与工程, 2022, 42 (2): 342. doi: 10.3969/j.issn.0258-0918.2022.02.015
26	李帅永, 毛维培, 程振华, 等. 基于VMD和K-SVD字典学习的供水管道泄漏振动信号压缩感知方法[J]. 仪器仪表学报, 2020, 41 (3): 49. doi: 10.19650/j.cnki.cjsi.J2005965
27	常浩伟, 庞春雷, 郭泽辉, 等. 基于自适应免疫算法的欺骗信号检测方法[J]. 系统工程与电子技术, 2022, 44 (8): 2419.
28	Jiang M Y, Luo M, Zhang C Y. Rolling bearing fault diagnosis based on acoustic-vibration data fusion and mode decomposition combined with the crested porcupine optimization algorithm[J]. Heliyon, 2024, 10 (22): es40351.
29	Meng Z, Liu J, Liu J B. An improved variational mode decomposition method based on spectrum reconstruction and segmentation and its application in rolling bearing fault diagnosis[J]. Digital Signal Processing, 2023, 141, 104161. doi: 10.1016/j.dsp.2023.104161
30	Ni Q, Ji J C, Feng K. A fault information-guided variational mode decomposition （FIVMD） method for rolling element bearings diagnosis[J]. Mechanical Systems and Signal Processing, 2022, 164, 108216. doi: 10.1016/j.ymssp.2021.108216
31	Zhang Y Q, Ren G Q, Wu D H. Rolling bearing fault diagnosis utilizing variational mode decomposition based fractal dimension estimation method[J]. Measurement, 2021, 181, 109614. doi: 10.1016/j.measurement.2021.109614
32	王磊, 张志勇, 胥辉旗, 等. 基于VMD分解和多域联合分布的雷达辐射源识别[J]. 系统工程与电子技术, 2023, 45 (8): 2479. doi: 10.12305/j.issn.1001-506X.2023.08.22
33	Gao Z, Liu Y, Wang Q J. Ensemble empirical mode decomposition energy moment entropy and enhanced long short-term memory for early fault prediction of bearing[J]. Measurement, 2022, 188, 110417. doi: 10.1016/j.measurement.2021.110417
34	Du H R, Wang J X, Qian W J. Rotating machinery fault diagnosis based on parameter-optimized variational mode decomposition[J]. Digital Signal Processing, 2024, 153, 104590. doi: 10.1016/j.dsp.2024.104590
35	Nassef M G A, Hussein T M, Mokhiamar O. An adaptive variational mode decomposition based on sailfish optimization algorithm and Gini index for fault identification in rolling bearings[J]. Measurement, 2021, 173, 108514. doi: 10.1016/j.measurement.2020.108514
36	Gong T K, Yuan X H, Yuan Y B. Application of tentative variational mode decomposition in fault feature detection of rolling element bearing[J]. Measurement, 2019, 135, 481. doi: 10.1016/j.measurement.2018.11.083
37	Zhang M, Jiang Z N, Feng K. Research on variational mode decomposition in rolling bearings fault diagnosis of the multistage centrifugal pump[J]. Mechanical Systems and Signal Processing, 2017, 93, 460. doi: 10.1016/j.ymssp.2017.02.013
38	Sato S, Kitamura T, Ando Y. Loudness of sharply （2068 dB/Octave） filtered noises in relation to the factors extracted from the autocorrelation function[J]. Journal of Sound and Vibration, 2002, 250 (1): 47. doi: 10.1006/jsvi.2001.3888
39	Tang M J, Liao Y H, He D. Rolling bearing diagnosis based on an unbiased-autocorrelation morphological filter method[J]. Measurement, 2022, 189, 110617. doi: 10.1016/j.measurement.2021.110617
40	Cheng Y, Wang Z W, Chen B Y. An improved complementary ensemble empirical mode decomposition with adaptive noise and its application to rolling element bearing fault diagnosis[J]. ISA Transactions, 2019, 91, 218. doi: 10.1016/j.isatra.2019.01.038
41	Xing Z, Yi C, Lin J H. Multi-component fault diagnosis of wheelset-bearing using shift-invariant impulsive dictionary matching pursuit and sparrow search algorithm[J]. Measurement, 2021, 178, 109375. doi: 10.1016/j.measurement.2021.109375
42	Randall R B, Antoni J. Why EMD and similar decompositions are of little benefit for bearing diagnostics[J]. Mechanical Systems and Signal Processing, 2023, 192, 110207. doi: 10.1016/j.ymssp.2023.110207
43	陈俊洵, 程龙生, 余慧, 等. 基于EMD-SVD与马田系统的复杂系统健康状态评估[J]. 系统工程与电子技术, 2017, 39 (7): 1542. doi: 10.3969/j.issn.1001-506X.2017.07.16
44	Martins D H, De-Lima A A, Gutiérrez R H R. Improved variational mode decomposition for combined imbalance-and-misalignment fault recognition and severity quantification[J]. Engineering Applications of Artificial Intelligence, 2023, 124, 106516. doi: 10.1016/j.engappai.2023.106516
45	Wang H L, Wu F, Zhang L. Application of variational mode decomposition optimized with improved whale optimization algorithm in bearing failure diagnosis[J]. Alexandria Engineering Journal, 2021, 60 (5): 4689. doi: 10.1016/j.aej.2021.03.034
46	Li Z X, Jiang Y, Guo Q, Hu C. Multi-dimensional variational mode decomposition for bearing-crack detection in wind turbines with large driving-speed variations[J]. Renewable Energy, 2018, 116, 55. doi: 10.1016/j.renene.2016.12.013
47	Chen X J, Yang Y M, Cui Z X. Vibration fault diagnosis of wind turbines based on variational mode decomposition and energy entropy[J]. Energy, 2019, 174, 1100. doi: 10.1016/j.energy.2019.03.057
48	Lin Y, Xiao M H, Liu H J. Gear fault diagnosis based on CS-improved variational mode decomposition and probabilistic neural network[J]. Measurement, 2022, 192, 110913. doi: 10.1016/j.measurement.2022.110913
49	Shi P M, Gao S M, Tan H. MGGSED-SSA: an improved sparse deconvolution method for rolling element bearing diagnosis[J]. Applied Acoustics, 2024, 220, 109960. doi: 10.1016/j.apacoust.2024.109960
50	Xu L, Cai D S, Shen W. Denoising method for fiber optic gyro measurement signal of face slab deflection of concrete face rockfill dam based on sparrow search algorithm and variational modal decomposition[J]. Sensors and Actuators A: Physical, 2021, 331, 112913. doi: 10.1016/j.sna.2021.112913

数据集	收敛迭代次数	适应度值	收敛速度提升/%
OR007@6_0	3	7.4010	80
OR007@6_2	5	7.3701	20
OR007@6_3	3	7.2740	57

参数	轴承直径/mm	滚子直径/mm	滚子数量	接触角度/（°）
尺寸	71.5	8.4	16	15.17

参数	轴承直径/mm	滚子直径/mm	滚子数量	接触角度/（°）
尺寸	39.80	7.92	8	0

数据集	理论频率/Hz	检测频率/Hz	误差/%	准确率/%	谐波可见性
西储1	107.31	108.00	0.64	99.36	显著
西储2	104.44	105.00	0.54	99.46	显著
西储3	102.88	102.00	1.71	98.29	显著
辛辛那提	236.40	235.60	0.34	99.66	显著
XJTU-SY	107.91	106.41	1.40	98.60	显著

方法	检测频率/Hz	误差/%	准确率/%
所提方法	235.6	0.34	99.66
PSO-VMD	240.3	1.62	98.38
GWO-VMD	244.1	3.15	96.85
SSA-VMD	228.7	3.26	96.74
普通VMD	224.6	4.99	95.01
EMD	220.6	6.68	93.32