基于SDG模型和模糊融合的故障诊断方法  

Fault Diagnosis Method Based on SDG Model and Fuzzy Fusion

作　　者：刘敏华[1] 萧德云[1] 

机构地区：[1]清华大学自动化系,北京100084

出　　处：《控制工程》2006年第1期10-14,共5页Control Engineering of China

基　　金：国家高技术研究发展计划(863计划)资助项目(2003AA412310)

摘　　要：将符号定向图SDG(Signed Directed Graph)节点偏差模糊化为定性模糊偏差,并从子节点向父节点对其进行数据融合,得到父节点故障源可能性,分析了该模糊融合传递信息的特性和处理各种发生故障的情形的优点。建立了基于SDG模型和模糊融合的故障诊断方法,包括SDG建模、故障源数检验和故障诊断等步骤,并以铝电解槽故障诊断为实例,说明了这种方法的有效性。The deviation of SDG node is fuzzed as qualitative fuzzy deviation, and the new deviations of child nodes are fused into the possibility of fault source of parent node with the analysis of the character of fault information transfer and the adventage of dealing with different malfunction cases. Then the fault diagnosis method based on SDG and fuzzy fusion is built, including SDG modeling, n-ary fault source test and fault diagnosis involved. An example of fault diagnosis on aluminum reduction cell shows the validity of the method.

关 键 词：符号定向图 模糊融合 故障诊断 铝电解槽 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]