|
Abstract:
This paper deals with the failure detection of manufacturing
systems. A learnable failure detection system has been proposed. Based
on the pattern recognition theory and fuzzy theory the decision problem
of failure detection has been solved for the case that two pattern can
not be separated clearly. And an evaluation method for failure detection
effectiveness has been proposed, which can be used for evaluation
effectiveness of failure detection of different random processed and
different failure grade. Finally, the effectiveness of failure detection
system is validate by computer simulation. The results of human
performance in monitoring are given in this paper as well.
Key words:
模式识别 机械加工 故障
国家自然科学基金资助项目. Received 198901, received in revised form 199102
|