Vision-assisted thermal monitoring system for CNC machine surveillance

Abstract

Fault diagnosis and predictive maintenance indicate an economic issue as an efficient technique is able to continuously monitor the concerned process and trigger an alarm when critical changes in key variables are detected. In this paper, we present a CNC machine surveillance system based on vision-assisted thermal monitoring approach. Firstly, a calibrated camera will detect and track the milling and transmit the position data to a host computer. Secondly, a laser built-in thermometer will continuously read the temperature of the milling by following the milling based on the position data. Finally, the host computer will generate an alarm signal when the temperature exceeds a pre-set threshold. © 2008 IEEE.