Fault Forecast Technology and Its Application in Predictive Maintenance
To realize fault forecast and predictive maintenance for complicated equipments, a new hybrid-inference fault forecast model integrated with dynamic fuzzy theory and Multi-Agent System (MAS) is firstly put forward. It combines forward inference and backward inference, different knowledge representations and inference mechanisms into a uniform inference framework, and resolve the puzzle of carrying out fault forecast for individual unit in certain conditions. The inference flow, weight coefficient matrix, dynamic fuzzy matrix and structure of multiAgent diagnosis federation are illustrated in detail. A candidate fault set including the probability of each fault symptom is provided by forward inference based on dynamic fuzzy theory. Then, the factual fault symptoms, the corresponding fault causations and components can be confirmed by inverted inference process based on multi-Agent synchronous inference framework.
Fault forecast predictive maintenance dynamic fuzzy inference MAS
Cao Lijun Hu Huibin Xu Maozu Qin Junqi
Shijiazhuang Mechanical Engineering College,Shijiazhuang,050003,China Shijiazhuang Mechanical Engineering College,Shijiazhuang,050003,China Shijiazhuang Army Command Coll
国际会议
厦门
英文
1257-1261
2010-05-22（万方平台首次上网日期，不代表论文的发表时间）