Novel Method for Vibration Sensor-Based Instantaneous Defect Frequency Estimation for Rolling Bearings Under...

Area of Science:

• Signal Processing
• Mechanical Engineering
• Vibration Analysis

Background:

• Estimating instantaneous frequencies (IF) in non-stationary signals is challenging, especially in noisy environments.
• Traditional time-frequency representations struggle to accurately extract true IF variations under strong background noise.
• Accurate IF estimation is crucial for diagnosing machinery health, such as identifying rolling bearing defects.

Purpose of the Study:

• To propose a novel instantaneous frequency estimation technology called multi-generalized demodulation transform (MGDT).
• To establish the theoretical foundations for the MGDT.
• To validate the effectiveness of MGDT through simulations and experimental data.

Main Methods:

• Development of the multi-generalized demodulation transform (MGDT) for IF estimation.
• Simulation of four types of modulated non-stationary signals with signal-to-noise ratios as low as -5 dB.
• Experimental application of MGDT to rolling bearing defect frequency analysis under various noise conditions.

Main Results:

• MGDT demonstrated accurate IF estimation for simulated non-stationary signals in highly noisy environments.
• Experimental results confirmed MGDT's effectiveness in identifying multiple defect frequency harmonics in rolling bearings.
• Quantitative analysis of instantaneous frequency estimation errors showed high performance of the proposed technology.

Conclusions:

• The multi-generalized demodulation transform is a novel and effective technology for instantaneous frequency estimation.
• MGDT overcomes limitations of traditional methods in noisy and non-stationary conditions.
• The technology shows significant promise for applications in condition monitoring and fault diagnosis.