Predictive Maintenance and Diagnostics of Machine Tools
Authors: Fabrizio Meo, Maxim Foursa, Sandor Kopacsi, Thomas Schlegel
Abstract
The objective of condition-based predictive maintenance applied to machine tools is the “measurement†of the slow degradation taking place in mechanical components in advance of the consequent failures and the evaluation of the residual life in order to better manage the total Life Cycle Cost. An automatic self-tuning of the system should accompany this procedure, in order to always optimize the machine tool performance as an optimal trade-off between surface quality and machining speed. These activities may be performed by intelligent software modules, communicating among them, with the machines, the sensors and the humans. The implementation of some of these modules is taking place in the frame of the project “Intelligent Networked Manufacturing System†(INT-MANUS).
Dear authors,
what kind of implementation on a machine tool do you suppose? Or have any?
Regards Pavel
The idea is that the software modules described, at least the Data Mining, Decision support and Tuning, may run in the environment of the Numerical Control of the machine.
Dear authors,
What dose that mean "running in the environment of the Numerical Control of the machine."?
Dose the software running in the CNC control system or an individual computer outside of the CNC machine?
Kind regards,
Xuedong
Yes, what does the enviroment mean? To my previous question: I ment, if you have already proven your system at a machine tool.
Thanks Pavel
Many CNC control systems today are PC based, and specifically in the case of a Fidia control, this is based on a PC running Windows XP, plus a real-time environment based on proprietary hardware and software. In this case the relative software is running on the PC (Windows) part of the control, and communicating with the real-time part through a dedicated API.
Yes, even though not in a uniform way. I mean that the Data Mining software was developed at the beginning, so we had time enough to test it on a significant number of machines. Since we were satisfied with the result, we decided that all machines manufactured by our company will run this software before being shipped, so we measure the indicators on all machines, and compare the values resulting from similar machines.
While the software modules for Decision Support and Tuning were developed later (and are still under development, even though we have a first version running now), so we had so far only enough time to make a few tests.
Dear authors,
Thank you very much for your contribution to IPROMS2008.
Could you please upload your video presentation similarly to what other authors have done in this session?
Kind regards,
Emmanuel