Bearing Fault Diagnosis Using a Hybrid Fuzzy V-Structure Fault Estimator Scheme

Area of Science:

• Mechanical Engineering
• Signal Processing
• Artificial Intelligence

Background:

• Bearings are vital motor components susceptible to faults that can cause system damage and economic losses.
• Early and accurate fault detection is crucial for preventing catastrophic failures and minimizing downtime.
• Existing fault diagnosis methods may struggle with oscillation and prediction accuracy under unknown conditions.

Purpose of the Study:

• To develop and evaluate a hybrid fuzzy V-structure fuzzy fault estimator for bearing fault diagnosis and crack size identification.
• To improve estimation and prediction accuracy for unknown bearing conditions.
• To enhance the robustness and stability of fault diagnosis algorithms.

Main Methods:

• A hybrid fuzzy V-structure fuzzy fault estimator was designed, combining fuzzy logic and V-structure approaches.
• A parallel fuzzy method was integrated to boost the robustness and stability of the V-structure algorithm.
• Data modeling utilized external autoregression error, a Laguerre filter, and support vector regression, with support vector machine for classification and crack size detection.

Main Results:

• The proposed hybrid fuzzy V-structure fault estimator demonstrated high effectiveness in diagnosing bearing faults using vibration signals.
• The method achieved an average accuracy of 98.75% for fault classification and 98% for crack size identification.
• The approach successfully differentiated between normal, ball, inner, and outer fault conditions.

Conclusions:

• The hybrid fuzzy V-structure fuzzy fault estimator offers a robust and accurate solution for bearing fault diagnosis and crack size identification.
• The integration of fuzzy logic and V-structure significantly reduces oscillation and improves stability.
• This advanced diagnostic technique holds potential for enhancing the reliability and economic efficiency of motor systems.