摘要
为了诊断并发故障,文中扩展了证据理论:将并发故障状态作为设备的运行状态引入分辨框Q中,视其为独立的运行状态,并根据不同证据体的差异程度,提出了两种不同的证据组合算法。然后,应用此两种不同的证据组合算法于并发故障诊断:第一种融合算法用来处理有相同焦点元素的证据体;第二种融合算法用来处理有不同焦点元素的证据体。该算法有利于充分利用各种证据体的有用信息,具有较强处理不确定性、不一致性证据体的能力。通过实验台转子轴承系统并发故障的诊断事例,证实了该算法的有效性。
A new information fusion approach based on extended Dempster-Shafer theory is proposed to diagnose simultaneous faults. Simultaneous faults are introduced in the frame of discernment and two different combination rules are established. The first combination rule fuses these bodies of evidence with same focal elements, the second combination rule fuses these bodies of evidence with different focal elements. Finally, a case study pertaining to the fault diagnosis for rotor-bearing systems is given, and by comparing and analyzing the diagnosis results based on the first combination rule and the second combination rule respectively, it is shown that the presented diagnosis approach is feasible and efficient.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2003年第5期112-115,共4页
Proceedings of the CSEE
关键词
发电机
转子
定子
信息融合
并发故障诊断
证据理论
人工神经网络
Electric power equipment
Information fusion
Dempster-shafer theory
Simultaneous fault
Fault diagnosis
