A Modified EMD Technique for Broken Rotor Bar Fault Detection in Induction Machines

Area of Science:

• Electrical Engineering
• Mechanical Engineering
• Industrial Monitoring

Background:

• Induction machines (IMs) are critical in industrial applications, and early detection of defects like broken rotor bars (BRBs) is vital for preventing performance degradation and ensuring safety.
• BRB faults can lead to significant issues including overheating, vibration, acoustic noise, and potential sparking, necessitating reliable detection methods.

Purpose of the Study:

• To propose and validate a modified empirical mode decomposition (MEMD) technique for accurate broken rotor bar (BRB) fault detection in induction machines.
• To utilize motor current signature analysis (MCSA) combined with MEMD for enhanced fault diagnosis.

Main Methods:

• Development of a smart sensor-based data acquisition (DAQ) system for wireless current signal collection.
• Application of a modified empirical mode decomposition (MEMD) technique involving correlation-based IMF selection and adaptive spectral analysis.
• Introduction of a novel reference function for analytical fault index generation and severity diagnosis.

Main Results:

• The proposed MEMD technique successfully identified BRB faults in induction machines.
• Experimental verification demonstrated the effectiveness of the MEMD approach in fault detection and severity assessment.
• The developed DAQ system facilitated efficient and wireless data collection for analysis.

Conclusions:

• The modified empirical mode decomposition (MEMD) technique offers a robust and effective solution for broken rotor bar (BRB) fault detection in induction machines.
• The integration of MEMD with motor current signature analysis provides a powerful tool for industrial condition monitoring.
• The study highlights the potential of advanced signal processing techniques for improving the reliability and safety of industrial machinery.