Nonlinear vibration feature extraction based on power spectrum envelope adaptive empirical Fourier decomposition

Area of Science:

• Marine Engineering
• Signal Processing
• Vibration Analysis

Background:

• Shipboard stabilized platforms (SSPs) require accurate vibration analysis for marine gravimetric surveys.
• Existing methods like Empirical Fourier Decomposition (EFD) struggle with noise and require manual parameter setting.
• Nonlinear and non-stationary vibration characteristics are crucial for effective isolation system design.

Purpose of the Study:

• To develop an adaptive and noise-robust method for extracting nonlinear vibration characteristics of SSPs.
• To improve the accuracy of signal decomposition compared to existing techniques.
• To enhance the design and vibration compensation of marine gravimetric survey isolation systems.

Main Methods:

• Proposed Power Spectrum Envelope Adaptive Empirical Fourier Decomposition (PSEEFD) for nonlinear signal analysis.
• Utilized mutual information for adaptive determination of modal decomposition.
• Implemented an improved power spectrum envelope segmentation to minimize noise interference.
• Reconstructed component signals using inverse fast Fourier transform to obtain instantaneous frequencies.

Main Results:

• PSEEFD demonstrated superior noise suppression compared to EMD, AFVMD, EWT, and EFD.
• The method effectively extracted nonlinear vibration characteristics from both simulated and measured SSP data.
• Adaptive segmentation based on mutual information improved decomposition accuracy.
• The power spectrum envelope method successfully reduced interference in spectrum segments.

Conclusions:

• PSEEFD offers a significant advancement in analyzing nonlinear vibration signals from SSPs.
• The method provides more accurate and reliable vibration characteristic extraction for marine applications.
• PSEEFD is a promising tool for enhancing the performance of marine gravimetric survey vibration isolation systems.