Sensor-Based Air-Gap Monitoring of Elevator Brakes via RMS-Envelope-Guided Transient Impact Extraction and RBF-SVM

Area of Science:

• Engineering
• Vibration Analysis
• Condition Monitoring

Background:

• Air-gap variation in elevator brakes is a critical indicator for brake condition monitoring.
• Analyzing vibration data from brakes is challenging due to mixed operating stages.
• Isolating short action-related impacts from long vibration recordings is difficult.

Purpose of the Study:

• To develop an engineering-oriented, sensor-based framework for elevator brake air-gap monitoring.
• To combine RMS-envelope-guided transient impact extraction with a classifier for accurate air-gap identification.
• To analyze the sensitivity of different vibration features to air-gap variations.

Main Methods:

• A triaxial accelerometer was mounted on the brake armature.
• Short-time RMS envelope was used to identify operating-state transition boundaries.
• Transient impact samples were extracted and characterized using time-domain, frequency-domain, and band-energy features.
• A Radial Basis Function Support Vector Machine (RBF-SVM) classifier was employed.

Main Results:

• Brake-release impact features demonstrated higher sensitivity to air-gap variation compared to engagement or combined features.
• The proposed RMS-IE-SVM method achieved high accuracy (up to 89.77%) in identifying air-gap states.
• Multi-axis vibration feature fusion (X+Y+Z) further improved cross-file generalization accuracy (89.63%).
• Time-domain features were found to provide dominant information, with band-energy features offering complementary insights.

Conclusions:

• The RMS-envelope-guided transient impact extraction framework provides an interpretable method for elevator brake air-gap identification under controlled conditions.
• Multi-axis vibration data enhances the generalization capability of the air-gap monitoring system.
• This approach contributes to more effective elevator brake condition monitoring and maintenance strategies.