基于改进的稀疏度自适应振动数据修复方法

PDF(1453 KB)

振动与冲击 ›› 2019, Vol. 38 ›› Issue (16) : 261-266.

论文

基于改进的稀疏度自适应振动数据修复方法

谢馨，王华庆，宋浏阳，李景乐，郝彦嵩

作者信息 +

A vibration data recovery method based on an modified sparsity adaptive algorithm

XIE Xin WANG Huaqing SONG Liuyang LI Jingle HAO Yansong

Author information +

文章历史 +

摘要

基于压缩感知的数据重构方法已用于解决信号采集中受损数据的修复问题，该算法首先需要已知数据稀疏度，而振动信号的稀疏度通常难以确定，增加了数据修复的难度。稀疏度自适应匹配追踪算法（SAMP）无需预估信号稀疏度，可用于受损数据修复，但SAMP算法的修复结果受终止条件影响较大，导致修复精度不高且效率较低。为此提出了基于终止准则改进的稀疏度自适应数据修复方法。首先基于振动信号波形特征和先验知识，选择适当的字典矩阵实现信号稀疏化；其次以单位矩阵为基础，根据数据的缺失模型构造观测矩阵；最后为了避免传统SAMP算法终止系数选取不当，导致支撑集引入错误原子的问题，采用改进的SAMP算法重构出完整信号，实现受损数据修复。通过仿真信号及轴承实测信号验证了方法有效性，且改进的SAMP算法在重构精度和运算效率上均有所提高。此外，改进的SAMP算法重构效果优于正交匹配追踪（OMP）与正则化正交匹配追踪（ROMP）。

Abstract

A data recovery method with compressed sensing (CS) theory was developed for the acquired signal. Since the sparsity of vibration signal is unknown which impedes its application, the sparsity adaptive matching pursuit (SAMP) algorithm was served for the reconstruction. However, the SAMP algorithm is greatly influenced by termination condition, which will lead to unsatisfied results. In this case, a modified SAMP algorithm based on termination criterion was proposed. Firstly, the dictionary matrix was selected according to the waveform characteristics and other priori knowledge. Then based on the unit matrix, the observation matrix can be constructed under the mission data model. Finally, to overcome the blind choice of the termination coefficient which will result in the support set with error atoms, the modified SAMP algorithm based on termination criterion was applied to recover the complete signal. The efficiency of the proposed method was validated by simulation signals and practical bearing signals, and the modified algorithm has better performance on reconstruction accuracy and computing efficiency. Besides that, the modified algorithm also outperforms orthogonal matching pursuit (OMP) and regularized orthogonal matching pursuit (ROMP).

导出引用

谢馨，王华庆，宋浏阳，李景乐,郝彦嵩 . 基于改进的稀疏度自适应振动数据修复方法[J]. 振动与冲击, 2019, 38(16): 261-266

XIE Xin WANG Huaqing SONG Liuyang LI Jingle HAO Yansong . A vibration data recovery method based on an modified sparsity adaptive algorithm[J]. Journal of Vibration and Shock, 2019, 38(16): 261-266

参考文献

[1] 刘畅, 伍星, 刘韬,等. 基于近似等距投影和支持向量机的滚动轴承故障诊断[J]. 振动与冲击, 2018, 37(5).
Liu Chang, Wu Xing, Liu Tao, et al. Fault diagnosis of rolling bearings based on near-isometric projection and support vector machine[J]. Journal of vibration and shock, 2018, 37(5).
[2] Cui L, Huang J, Zhang F. Quantitative and Localization Diagnosis of a Defective Ball Bearing Based on Vertical-Horizontal Synchronization Signal Analysis[J]. IEEE Transactions on Industrial Electronics, 2017, 64(11): 8695-8706.
[3] Cai G, Chen X, He Z. Sparsity-enabled signal decomposition using tunable Q-factor wavelet transform for fault feature extraction of gearbox[J]. Mechanical Systems & Signal Processing, 2013, 41(1-2):34-53.
[4] Perepu S K, Tangirala A K. Reconstruction of missing data using compressed sensing techniques with adaptive dictionary[J]. Journal of Process Control, 2016, 47:175-190.
[5] 张新鹏, 胡茑庆, 程哲,等. 基于压缩感知的振动数据修复方法[J]. 物理学报, 2014, 63(20):115-124.
ZHANG Xin-peng, HU Niao-qing, CHENG Zhe, et al. Vibration data recovery based on compressed sensing[J]. Acta Physica Sinica. 2014, 63(20):115-124.
[6] 周亚同, 王丽莉, 唐红梅. 基于压缩感知的稀疏度自适应图像修复[J]. 铁道学报, 2014(9):52-59.
ZHOU Ya-tong, WANG Li-li, TANG Hong-mei. Sparsity adaptive algorithm for image inpainting Based on compressive sensing[J]. Journal of the China Railway Society, 2014(9):52-59.
[7] 余发军, 周凤星, 严保康. 基于字典学习的轴承早期故障稀疏特征提取[J]. 振动与冲击, 2016, 35(6):181-186.
YU Fa-jun, ZHOU Feng-xing, YAN Bao-kang. Bearing initial fault feature extraction via representation based on dictionary learning[J]. Journal of vibration and shock, 2016, 35(6): 181-186.
[8] 杨真真, 杨震, 孙林慧. 信号压缩重构的正交匹配追踪类算法综述[J]. 信号处理, 2013, 29(4):486-496.
YANG Zhen-zhen, YANG Zhen, SUN Lin-hui. A Survey on Orthogonal Matching Pursuit Type Algorithms for Signal Compression and Reconstruction[J]. Journal of signal processing, 2013, 29(4): 486-496
.

[9] 高洪波, 刘杰, 李允公. 基于实测冲击响应字典稀疏表示的齿轮系统侧隙故障特征提取[J]. 振动与冲击, 2017, 36(11):86-91.
Gao Hongbo, Liu Jie, Li Yungong. Backlash fault feature extraction of spur gear systems based on sparse representation dictionary of measured impulse responses [J]. Journal of vibration and shock, 2017, 36(11) :86-91.
[10] Cui L, Wu N, Ma C, et al. Quantitative fault analysis of roller bearings based on a novel matching pursuit method with a new step-impulse dictionary[J]. Mechanical Systems & Signal Processing, 2016, 68.
[11] Do T T, Lu G, Nguyen N, et al. Sparsity adaptive matching pursuit algorithm for practical compressed sensing[C]. Signals, Systems and Computers, 2008, Asilomar Conference on. IEEE, 2009:581-587.
[12] Wei Y, Lu Z, Yuan G, et al. Sparsity Adaptive Matching Pursuit Detection Algorithm Based on Compressed Sensing for Radar Signals[J]. Sensors, 2017, 17(5):1120.
[13] Du Z, Chen X, Zhang H. Feature identification with compressive measurements for machine fault diagnosis[C] Instrumentation and Measurement Technology Conference. IEEE, 2015:977-987.
[14] 周俊, 伍星, 迟毅林,等. 基于改进频域压缩感知的轴承复合故障欠定盲提取[J]. 振动与冲击, 2015, 34(14):123-128.
ZHOU Jun, WU Xing, CHI Yilin. Underdetermained blind extraction for bearing complex failures based on improved frequency damain compressive sensing[J].Journal of vibration and shock,2015, 34(14):123-128.
[15] Wang H, Ke Y, Song L, et al. A Sparsity-Promoted Decomposition for Compressed Fault Diagnosis of Roller Bearings[J]. Sensors, 2016, 16(9).
[16] 苗中华, 周广兴, 刘海宁,等. 基于稀疏编码的振动信号特征提取算法与实验研究[J]. 振动与冲击, 2014, 33(15): 76-81.
Miao Zhonghua, Zhou Guangxing, Liu Haining, et al. Tests and feature extraction algorithm of vibration signals based on sparse coding[J]. Journal of vibration and shock, 2014, 33(15):76-81.
[17] Sancetta A. Greedy algorithms for prediction[J]. Bernoulli, 2016, 22(2):1227-1277.