Application Combining VMD and ResNet101 in Intelligent Diagnosis of Motor Faults

Area of Science:

• Engineering
• Artificial Intelligence
• Signal Processing

Background:

• Motor failures pose significant risks to critical machinery like wind turbines and electric vehicles.
• Effective fault diagnosis is crucial for ensuring the safe and reliable operation of motor-driven equipment.
• Existing methods often require separate steps for signal analysis and feature extraction.

Purpose of the Study:

• To develop an end-to-end intelligent fault diagnosis system for motors.
• To integrate signal decomposition, feature extraction, and health status recognition into a unified framework.
• To improve the accuracy and efficiency of motor fault detection.

Main Methods:

• Utilized Variational Mode Decomposition (VMD) for adaptive signal processing and feature extraction from motor vibration signals.
• Employed a deep learning model, specifically Residual Neural Network 101 (ResNet101), for robust feature learning and classification.
• Developed a framework for unified pre-analysis, feature extraction, and health status recognition.

Main Results:

• The proposed system achieved a prediction accuracy rate of 94% in identifying six common motor faults.
• Demonstrated the effectiveness of combining VMD and ResNet101 for motor fault diagnosis.
• Validated the system's performance using a dataset from the Federal University of Rio de Janeiro (UFRJ).

Conclusions:

• The integrated VMD-ResNet101 system offers a highly effective and accurate method for intelligent motor fault diagnosis.
• This approach provides a promising solution for enhancing the reliability and safety of mechanical equipment.
• The system has broad applicability in fault diagnosis engineering for various industrial applications.