A novel solution for improved performance of Time-frequency concentration. (15th February 2023)

• Record Type: Journal Article
• Authors: Guo, Juan; Hao, Guocheng; Yu, Jiantao; Wang, Panpan; Jin, Yarui
• Is Part Of: Mechanical systems and signal processing. Volume 185(2023)
• Publication Date: 2023-02-15
• Subjects: Non-stationary signals -- short-time fractional Fourier transform -- Synchroextracting transform -- Time–frequency concentration -- Variational mode decomposition; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2022.109784
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A novel VSSTFrFT method is proposed to improve the concentration of time-frequency distribution. Using the idea of optimal rotation order in FrFT to adaptively set the number of modal decomposition in VMD. SET is utilized to extract the coefficient of ridgeline position, improving the time-frequency concentration. The robustness and concentration performance of the proposed method are verified by simulation signals with different components. The effectiveness of the VSSTFrFT method is verified by the data from the cantilever hammering modal signal, bat echolocation chirp signal and bearing fault signals. Abstract: To improve the time–frequency (TF) concentration performance of the short-time fractional Fourier transform (STFrFT), and solve the multi-order matching problem of multi-component signals, this paper introduces a new algorithm that is referred to as the Variational mode decomposition-short-time fractional Fourier transform- Synchroextracting transform (VSSTFrFT). This work employs the Variational mode decomposition (VMD) algorithm to decompose multi-component signals into single-component sets, then the STFrFT algorithm matches the optimal rotation order of each component separately to solve the multi-order matching problem. Finally, this paper utilizes the Synchroextracting transform (SET) to extract the TF coefficient of ridgeline position in STFrFT distribution, improving the concentration performance. For different types of signals, the VSSTFrFT algorithm can get a higher TF concentration than the traditional TFA methods. In the application of engineering measured data, the VSSTFrFT algorithm can extract the modalities of the signals and display the frequency curve clearly, which can be used for structural complexity and time-varying characteristics analysis in practical engineering applications.