Original Date: 11/03/1996
Revision Date: 01/18/2007

In conjunction with a Cooperative Research and Development Agreement (CRADA) program, Oak Ridge is developing a predictive maintenance system for predicting machine tool maintenance needs and performance degradation during operation. The system will allow staff to schedule maintenance service at convenient times and reduce productivity losses associated with unexpected machine failures.

The predictive maintenance system uses sensors such as accelerometers, electrical current sensors, thermocouples, resistive temperature detectors, and pressure transducers to gather data from the machine tools during operations. LabView data acquisition software, standard signal processing algorithms, and electrical signature analysis (ESA) are also used to collect and analyze real-time data. By collecting and analyzing sensor data in real-time, the system can predict degradation trends and eliminate run-to-failure and time-directed maintenance methods.

Oak Ridge has assembled the predictive maintenance system and developed the graphical user interface needed for displaying the recorded parameters. The next task involves running a battery of tests. If Phase I is successful, the methodology and algorithms behind this system will be considered for integration into the machine tool controller. Possible benefits of predictive diagnostics include increasing machine availability and decreasing maintenance costs.

For more information see the Point of Contact for this survey.