Operation Status Tracking for Legacy Manufacturing Systems via Vibration Analysis

Abstract

Tracking the status of manufacturing systems is important for analyzing the performance of a manufacturing process. Unfortunately, legacy manufacturing systems are technologies from the yesteryears which have no Internet connectivity and very often are not programmable. Gathering operational information of such systems is often done manually with poor temporal resolution. This work proposes an Internet-of-things (IoT) approach that uses vibration sensors to track the operation status of legacy manufacturing systems. One of the challenges of using vibration data is to identify the meaning of the vibration without prior knowledge of the vibration profile and without the privilege to interrupt the manufacturing process. Although there are many existing works that capture and analyze vibration data, these existing works very often only focus on fault diagnosis and prognosis. Our work focuses on using the vibration data to monitor the operation status of a manufacturing machine. Experimental results show that the proposed vibration analysis method is able to track the operation status of a machine with more than 90% accuracy, in the worst case with 90.2% and standard uncertainty of 3.6%.