摘要
针对传统汽轮机转子故障诊断方法中存在精度不高的问题,引入了证据理论和灰色神经网络相结合的故障诊断方法。该方法首先根据所获取的故障特征向量,用灰色建模方法进行累加处理,以增强数据的规律性;然后经BP神经网络进行局部诊断,以获得彼此独立的证据;最后采用证据理论对各证据进行融合。试验结果表明,该方法具有较好的故障模式识别能力,适用于汽轮机转子的故障诊断,具有一定的应用价值。
Aiming at the demerit of low accuracy of traditional methods for turbine rotor fault diagnosis, the method of combining evidence theory and gray neural network is introduced. Firstly, according to the acquired eigenvectors of faults, the gray modeling method is applied in cumulative processing to enhance the regularity of the data. In addition, BP neural network is used to carry on local diagnosis and acquire independent evidence of each other, and finally the evidence theory is adopted to integrate various evidences. The experimental results show that the method possesses good capability to identify failure mode, it is suitable for fault diagnosis of the turbine rotor, and providing certain applicable value.
出处
《自动化仪表》
CAS
北大核心
2012年第11期12-14,共3页
Process Automation Instrumentation
