OLD (BETTER)
Computer
integrated manufacturing technology can be well-served by a predictive
monitoring system that would prevent a large number of sensors from
overwhelming the electronic data monitoring system or a human operator. The
essence of the method is to select only a few of the many sensors in the system
for monitoring at a given time and to set alarm levels of the selected sensor
outputs to reflect the limit of expected normal operation at the given time.
The method is intended for use in highly instrumented system that includes many
interfacing components and subsystems - for example, an advanced aircraft, an
environmental chamber, a chemical processing plant, or a machining work cell.
The
predictive monitoring method would be implemented in a computer system running
artificial intelligence software, tentatively named PREMON. The predictive
monitoring system would include threee modules: a casual simulator, a sensor
planner, a sensor interpreter.
(S.
Soloman, Sensors and Control Systems in Manufacturing, pp.314-315)
NEW
Quick detection of
process faults is a very important requirement of dynamic monitoring. A
predictive monitoring strategy for continuous processes is proposed here. In
this predictive monitoring approach, a data window containing predicted future
data is projected onto the low dimensional score space developed from the
reference set. The length of the data window is q and the predictive horizon
within this window is h, where h<q. A score and an SPE can be calculated for
this window coresponding to time t + h, where t is the current time. In this
way, predictive monitoring is carried out.
(Cornelius
T. Leondes, Computed Aided and Integrated Manufacturing Systems: Computer
techniques, pg 213)
No comments:
Post a Comment