Related Experiment Video
Updated: Aug 9, 2025
09:30
Super-resolution Imaging of Neuronal Dense-core Vesicles
Published on: July 2, 2014
9.8K
A Fault Feature Extraction Method Based on Improved VMD Multi-Scale Dispersion Entropy and TVD-CYCBD
Jingzong Yang1, Chengjiang Zhou2, Xuefeng Li3
1School of Dig Data, Baoshan University, Baoshan 678000, China.
Entropy (Basel, Switzerland)
|February 25, 2023
Summary
This study introduces an advanced method for extracting fault features from mechanical equipment signals. The technique effectively identifies fault characteristics even in noisy environments, improving machinery diagnostics.
Area of Science:
- Mechanical Engineering
- Signal Processing
- Fault Diagnosis
Background:
- Mechanical equipment faults are challenging to detect due to signal noise.
- Extracting fault features is crucial for predictive maintenance and operational safety.
Purpose of the Study:
- To develop an effective fault feature extraction method for mechanical equipment.
- To address the challenge of submerged fault signals in noisy industrial environments.
Main Methods:
- Proposed a novel method combining Variational Mode Decomposition (VMD) optimized by the Marine Predator Algorithm (MPA), multi-scale dispersion entropy, and TVD-CYCBD.
- MPA optimized VMD parameters for signal decomposition.
- TVD was used for denoising, followed by CYCBD filtering and envelope demodulation.
Main Results:
- The method successfully decomposed and filtered fault signals.
- Envelope spectrum analysis revealed clear multiple frequency-doubling peaks with minimal interference.
- Verified effectiveness through simulation and actual fault signal experiments.
Conclusions:
- The proposed fault feature extraction method demonstrates superior performance in identifying fault characteristics.
- This technique enhances the reliability of fault diagnosis in complex industrial settings.
