摘要
在凝汽器故障诊断的研究历史中,人工神经网络有着举足轻重的作用。但由于单一神经网络自身缺点的存在,使得故障诊断的准确率不高。针对这一问题,使用BP神经网络和Elman神经网络对故障样本进行训练仿真,并将诊断结果借助于D-S证据理论进行融合。结果表明:多神经网络与D-S证据理论融合的方法可以很好的进行故障剥离,诊断出最终故障,从而提高凝汽器故障诊断的准确性。
The artificial Neural Network is playing a great role in research of condenser fault diagnosis in history. The precision rate of fault diagnosis is not high since the personal defect of single neural network. BP neural network and Elman neural network are firstly used to train and simulate the same fault examples in this paper and the diagnostic results are fused with the help of the D-S evidence theory. The results show that the effect of eliminating doubtful faults is very well base on the fusion of multi-neural networks and D-S evidence theory, which distinguish the fault in the end and improve the accuracy of the condenser fault diagnosis.
出处
《汽轮机技术》
北大核心
2016年第2期122-124,121,共4页
Turbine Technology
基金
吉林省科技发展计划资助项目(20140204040SF)
