基于EEMD和SOM神经网络的水电机组故障诊断
发布时间：2018-09-08 16:06
【摘要】：针对水电机组振动信号的非平稳性和特殊性,提出一种基于集合经验模态分解(EEMD)的奇异谱熵和自组织特征映射网络(SOM)相结合的故障诊断方法。首先采用EEMD对振动信号进行分解,得到本征模态函数(IMF);随后进行奇异谱分解,得到反映振动信号的动态特征向量——奇异谱熵;最后将得到的特征向量输入经过训练的SOM神经网络中进行故障自动识别。结果表明:该方法可以准确地提取机组故障特征,具有更高的识别精度和更快的计算速度。
[Abstract]:In view of the non-stationarity and particularity of the vibration signals of hydroelectric generating units, a fault diagnosis method based on the singular spectral entropy and the self-organizing feature mapping network (SOM) based on the set empirical mode decomposition (EEMD) is proposed. Firstly, the vibration signal is decomposed by EEMD, then the eigenmode function (IMF); is decomposed by singular spectrum, and the dynamic eigenvector-singular spectral entropy is obtained. Finally, the obtained eigenvector is input to the trained SOM neural network for automatic fault identification. The results show that this method can accurately extract the fault features of the unit and has higher recognition accuracy and faster calculation speed.
【作者单位】： 西安理工大学水利水电学院;
【基金】：国家自然科学基金(51209172)
【分类号】：TV738
[Abstract]:In view of the non-stationarity and particularity of the vibration signals of hydroelectric generating units, a fault diagnosis method based on the singular spectral entropy and the self-organizing feature mapping network (SOM) based on the set empirical mode decomposition (EEMD) is proposed. Firstly, the vibration signal is decomposed by EEMD, then the eigenmode function (IMF); is decomposed by singular spectrum, and the dynamic eigenvector-singular spectral entropy is obtained. Finally, the obtained eigenvector is input to the trained SOM neural network for automatic fault identification. The results show that this method can accurately extract the fault features of the unit and has higher recognition accuracy and faster calculation speed.
【作者单位】： 西安理工大学水利水电学院;
【基金】：国家自然科学基金(51209172)
【分类号】：TV738
【参考文献】
相关期刊论文 前10条
1 安周鹏;肖志怀;陈宇凡;孙召辉;唐卫东;;小波包能量谱和功率谱分析在水电机组故障诊断中的应用[J];水力发电学报;2015年06期
2 薛浩然;张珂珩;李斌;彭晨辉;;基于布谷鸟算法和支持向量机的变压器故障诊断[J];电力系统保护与控制;2015年08期
3 王军辉;贾嵘;谭泊;;基于EEMD和模糊C均值聚类的风电机组齿轮箱故障诊断[J];太阳能学报;2015年02期
4 王贺;胡志坚;张翌晖;李晨;杨楠;王战胜;;基于聚类经验模态分解和最小二乘支持向量机的短期风速组合预测[J];电工技术学报;2014年04期
5 朱文龙;周建中;肖剑;肖汉;李超顺;;独立分量分析 经验模态分解特征提取在水电机组振动信号中的应用[J];中国电机工程学报;2013年29期
6 王贺;胡志坚;陈珍;仉梦林;贺建波;李晨;;基于集合经验模态分解和小波神经网络的短期风功率组合预测[J];电工技术学报;2013年09期
7 苏立;南海鹏;余向阳;吴罗长;王瑾;;基于改进阈值函数的小波降噪分析在水电机组振动信号中的应用[J];水力发电学报;2012年03期
8 苏立;南海鹏;余向阳;吴罗长;王瑾;;基于集合经验模态分解与Wigner-Ville分布的水轮机振动特征信号的分析研究[J];水力发电学报;2012年02期
9 罗兴，
本文编号：2231012
本文链接：https://www.wllwen.com/kejilunwen/shuiwenshuili/2231012.html
