Bearing Fault Diagnosis Based on Energy Spectrum Statistics and Modified Mayfly Optimization Algorithm.
Yuhu Liu1,2, Yi Chai1,2, Bowen Liu1,2
1College of Automation, Chongqing University, Chongqing 400044, China.
Sensors (Basel, Switzerland)
|April 3, 2021
Summary
A new method, the initial center frequency-guided filter (ICFGF), accurately diagnoses bearing faults by selecting optimal frequency bands. This technique effectively suppresses random impulse interference for improved fault detection.
Area of Science:
- Mechanical Engineering
- Signal Processing
- Condition Monitoring
Background:
- Bearing faults are critical in rotating machinery, leading to equipment failure.
- Accurate fault diagnosis requires effective signal processing to isolate fault signatures.
- Existing methods struggle with random impulse interference, hindering reliable diagnosis.
Purpose of the Study:
- To propose a novel resonance demodulation frequency band selection method for bearing fault diagnosis.
- To enhance the robustness against random impulse interference.
- To improve the accuracy of bearing fault feature extraction.
Main Methods:
- The initial center frequency-guided filter (ICFGF) method is introduced.
- Step 1: Variance statistic index to determine fault impulse center frequency and suppress interference.
- Step 2: Modified mayfly optimization algorithm (MMA) for optimal frequency band search, followed by squared envelope spectrum analysis.
Main Results:
- The ICFGF method effectively extracts bearing fault features from outer and ball fault signals.
- Demonstrated superior performance in resisting random impulse interference.
- Outperformed other methods like fast kurtogram and ensemble empirical mode decomposition in accuracy.
Conclusions:
- The ICFGF method is a robust and accurate technique for bearing fault diagnosis.
- It offers significant advantages in handling noisy signals with random impulses.
- The proposed method enhances condition monitoring capabilities for rotating machinery.
