Broken Rotor Bar Detection in Induction Motors through Contrast Estimation

Area of Science:

• Electrical Engineering
• Industrial Monitoring
• Fault Diagnosis

Background:

• Induction motors (IM) are critical industrial components requiring continuous monitoring for fault detection.
• Broken rotor bars (BRBs) are common but challenging faults to detect, often necessitating specialized techniques.
• Existing fault detection methods primarily focus on the IM startup transient, limiting their applicability.

Purpose of the Study:

• To develop an efficient fault detection technique for induction motors operating in a steady state.
• To introduce a novel methodology for identifying broken rotor bars (BRBs) using motor current signal analysis.
• To adapt contrast estimation, typically used in image processing, for diagnosing IM conditions.

Main Methods:

• Motor current signal analysis was employed to diagnose the condition of induction motors.
• Contrast estimation, specifically Unser and Tamura definitions, was applied to motor current signals.
• The methodology was tested for detecting healthy, one broken bar (1BB), and two broken bars (2BB) conditions.

Main Results:

• The proposed methodology successfully identified and classified IM operational conditions during steady state.
• Unser and Tamura contrast definitions proved to be effective indicators for fault diagnosis.
• High certainty was achieved in distinguishing between healthy, 1BB, and 2BB states.

Conclusions:

• The novel methodology offers a reliable approach for detecting broken rotor bars in induction motors during steady-state operation.
• Contrast estimation presents a viable, previously unexplored tool for induction motor fault diagnosis.
• This technique enhances the ability to monitor industrial processes and prevent production interruptions.