摘要
汽轮机通流部分运行时会出现各种故障,有些故障具有隐秘性,不易察觉,有必要对其进行监测与诊断。将工程应用中效果较好的自适应模糊推理神经网络(ANFIS)与自组织特征映射(SOM)网络的相结合,集二者之所长,建立一种新型ANFIS—SOM神经网络,并以电厂运行故障数据做了计算分析。结果表明:基于新网络的故障诊断效果得到极大的提高,为汽轮机故障诊断提供了新思路。
There are various faults in the operation of the turbine flow,and some faults are hidden and difficult to detect and need to be monitored and diagnosed.The ANFIS-SOM neural network is established by combining the adaptive network-based fuzzy inference system （ANFIS） and the self-organizing feature map （SOM） network in the engineering application.The power plant operation fault data is calculated and analyzed.The results show that： The fault diagnosis effect based on the new network has been deeply improved,which provides a new idea for fault diagnosis of steam turbine.
作者
王惠杰
张家宁
赵立坤
王雷雨
WANG Hui - jie ZHANG Jia - ning ZHAO Li -kun WANG Lei - yu(School of Energy, Power and Mechanical Engineering, North China Electric Power University, Baoding 071003, Hebei Province, China)
出处
《应用能源技术》
2017年第9期1-7,共7页
Applied Energy Technology
