A parameter-adaptive VMD method based on grasshopper optimization algorithm to analyze vibration signals from rotating machinery. (August 2018)

• Record Type: Journal Article
• Authors: Zhang, Xin; Miao, Qiang; Zhang, Heng; Wang, Lei
• Is Part Of: Mechanical systems and signal processing. Volume 108(2018)
• Page Start: 58
• Page End: 72
• Publication Date: 2018-08
• Subjects: Vibration signal analysis -- Fault diagnosis -- Variational mode decomposition -- Grasshopper optimization algorithm -- Weighted kurtosis index -- Parameter adaptive estimation; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2017.11.029
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A parameter-adaptive VMD based on GOA for vibration signal analysis is proposed. The weighted kurtosis index is defined as optimization objective function. GOA is used to search for the optimal VMD decomposition parameters. Case studies demonstrate effectiveness and advantages of the proposed method. Abstract: The mode number and mode frequency bandwidth control parameter (or quadratic penalty term) have significant effects on the decomposition results of the variational mode decomposition (VMD) method. In the conventional VMD method, the values of decomposition parameters are given in advance, which makes it difficult to achieve satisfactory analysis results. To address this issue, this paper proposes a parameter-adaptive VMD method based on grasshopper optimization algorithm (GOA) to analyze vibration signals from rotating machinery. In this method, the optimal mode number and mode frequency bandwidth control parameter that match with the analyzed vibration signal can be estimated adaptively. Firstly, a measurement index termed weighted kurtosis index is constructed by using kurtosis index and correlation coefficient. Then, the VMD parameters are optimized by the GOA algorithm using the maximum weighted kurtosis index as optimization objective. Finally, fault features can be extracted by analyzing the sensitive mode with maximum weighted kurtosis index. Two case studies demonstrate that the proposed method is effective to analyze machinery vibration signal for fault diagnosis. Moreover, comparisons with the conventional fixed-parameter VMD method and the well-known fast kurtogram method highlight the advantages of the proposed method.