In the process of equipment fault evolution, equipment's attrition probability is different, namely selective attrition may be produced. This paper describes one method based on selective attrition, which achieves the aim of forecasting its optimal maintenance time by analyzing and handling the data of monitoring equipment. There are mainly two steps: first, get each part’s probability value of selective attrition under the current condition in use of association rule algorithm, then take the obtained probability value as the input, and get the optimal maintenance time through neural networks modeling. This method employs the real-time and dynamic decision-making method and adjusts the optimal maintenance time in real time according to the information in the equipment’s operating process, making it more conform to the actual condition. The feasibility of this method is showed by a simulation example.
[1]
Qing-Hui Wang,et al.
A rough set-based fault ranking prototype system for fault diagnosis
,
2004,
Eng. Appl. Artif. Intell..
[2]
Alex Pentland,et al.
Probabilistic Visual Learning for Object Representation
,
1997,
IEEE Trans. Pattern Anal. Mach. Intell..
[3]
W. J. Wang,et al.
THE APPLICATION OF PSEUDO-PHASE PORTRAIT IN MACHINE CONDITION MONITORING
,
2003
.
[4]
Tomasz Imielinski,et al.
Database Mining: A Performance Perspective
,
1993,
IEEE Trans. Knowl. Data Eng..
[5]
Liangsheng Qu,et al.
Fault diagnosis using Rough Sets Theory
,
2000
.