基于信息融合技术的汽轮机转子故障诊断  

FAULT DIAGNOSIS FOR TURBINE ROTOR BASED ON INFORMATION FUSION TECHNOLOGY

作　　者：熊伟[1] 程加堂[1] 艾莉[1] 

机构地区：[1]红河学院工学院,云南蒙自661199

出　　处：《热力发电》2012年第8期62-64,共3页Thermal Power Generation

摘　　要：选取蚁群神经网络作为汽轮机转子故障诊断的初级模块,采用证据理论混合算法的故障诊断方法,对汽轮机转子的局部故障进行诊断,并将诊断结果作为证据体利用证据理论将各证据体进行合成,计算它们的基本可信任分配函数,从而判定故障及其类型。以汽轮机转子x、y方向的不平衡故障为例进行诊断,结果表明该方法可有效提高诊断的可信度,减少诊断的不确定性。Local fault diagnosis of turbine rotor was conducted by fault diagnosis method using evidence theory with hybrid algorithm,combined with ant colony algorithm-neural network model,and the results were considered as the evidence.Then the evidence theory was employed to fuse the evidences and calculate their basic trustable assignment function,so as to determine the fault and its type.Imbalance fault in x,y direction on the turbine rotor was diagnosed.Results show that,this method can efficiently enhance the reliability and decrease the uncertainty of the diagnosis.

关 键 词：汽轮机转子 信息融合 证据理论 蚁群神经网络 故障诊断 不平衡 

分 类 号：TK268.1[动力工程及工程热物理—动力机械及工程] TP306+.3[自动化与计算机技术—计算机系统结构]