This study focuses on the design and development of a condition monitoring system for stretch wrapping machines using accelerometer-based measurements. The purpose is to detect defects on the wheels of the wrapping carriage to enhance machine reliability and operational efficiency. A critical aspect of the study involved determining the optimal placement of accelerometers to maximize sensitivity to wheel defects. The investigation involved extensive testing under controlled conditions with several combinations of working conditions, various artificially induced defects and sensor locations. Data collected from the accelerometers were processed and analyzed using multiple statistical indicators, including bearing-related techniques and cyclostationary approaches. These analyses enabled the identification of key patterns and thresholds indicative of wheel defects. Preliminary results demonstrate the feasibility of detecting early-stage anomalies through accelerometer data and highlight the relevance of sensor positioning in capturing meaningful signals. The ongoing phase of the project focuses on refining the system’s accuracy and robustness through additional testing and refining of the data processing procedure. Future developments aim to integrate the system into machine control units for automated defect detection and alerts.
A VIBRATION BASED HEALTH MONITORING SYSTEM FOR A STRETCH WRAPPING MACHINE
Ilaria RizioliPrimo
;Giulia CristoforiSecondo
;Mattia Battarra
;Emiliano MucchiUltimo
2025
Abstract
This study focuses on the design and development of a condition monitoring system for stretch wrapping machines using accelerometer-based measurements. The purpose is to detect defects on the wheels of the wrapping carriage to enhance machine reliability and operational efficiency. A critical aspect of the study involved determining the optimal placement of accelerometers to maximize sensitivity to wheel defects. The investigation involved extensive testing under controlled conditions with several combinations of working conditions, various artificially induced defects and sensor locations. Data collected from the accelerometers were processed and analyzed using multiple statistical indicators, including bearing-related techniques and cyclostationary approaches. These analyses enabled the identification of key patterns and thresholds indicative of wheel defects. Preliminary results demonstrate the feasibility of detecting early-stage anomalies through accelerometer data and highlight the relevance of sensor positioning in capturing meaningful signals. The ongoing phase of the project focuses on refining the system’s accuracy and robustness through additional testing and refining of the data processing procedure. Future developments aim to integrate the system into machine control units for automated defect detection and alerts.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


