Quantitative and Localization Fault Diagnosis Method of Rolling Bearing Based on Quantitative Mapping Model

Area of Science:

• Mechanical Engineering
• Vibration Analysis
• Condition Monitoring

Background:

• Accurate diagnosis of rolling bearing faults is crucial for machinery reliability.
• Existing methods struggle with precise fault size and location determination.
• Vibration signals contain complex information influenced by fault characteristics.

Purpose of the Study:

• To propose a novel quantitative and localization fault diagnosis method for rolling bearings.
• To establish a quantitative mapping model (QMM) for predicting fault size.
• To achieve accurate localization of bearing faults.

Main Methods:

• Development of a quantitative mapping model (QMM) based on vibration signal analysis.
• Utilizing multiscale permutation entropy (MPE) to quantify signal complexity related to fault size.
• Employing multiscale morphological filtering (MMF) for feature extraction and QMM accuracy enhancement.
• Establishing linear and nonlinear QMMs through regression functions.

Main Results:

• The proposed QMM effectively predicts rolling bearing fault size.
• The method successfully achieves localization diagnosis of faults.
• MMF improves the accuracy of fault size prediction through enhanced feature extraction.
• The QMM demonstrates significant impact on quantitative fault size prediction and localization.

Conclusions:

• The novel QMM-based method provides accurate quantitative fault size prediction for rolling bearings.
• The method enables effective localization diagnosis of rolling bearing faults.
• The integration of MPE and MMF enhances the diagnostic capabilities for bearing health monitoring.