基于小波包和样本熵的水泵机组振动特征提取

宋媛 彭利鸿 赖冠文 张嘉勋 肖志怀 宋丽波

PDF(2553 KB)
中国农村水利水电 ›› 2017 ›› (3) : 146-152.
水电建设

基于小波包和样本熵的水泵机组振动特征提取

  • 宋媛1,彭利鸿1,赖冠文2,张嘉勋2,肖志怀1,宋丽波3
作者信息 +

Vibration Feature Extraction of Pump Unit Based on Wavelet Packet and Sample Entropy

  • SONG Yuan1,PENG Li-hong1,LAIGuan-wen2,ZHANG Jia-xun2,XIAO Zhi-huai1,SONG Li-bo3#br#
Author information +
稿件信息 +

摘要

目前大部分大型水泵机组安装有状态监测系统,但如何从海量的状态监测数据中提取出机组故障特征仍是水泵机组故障诊断的一大难点和热点。本文提出了一种基于小波包和样本熵的水泵机组振动特征提取方法,该方法首先通过小波包变换对水泵机组振动信号进行分层分解,得到小波包频带系数,再结合样本熵算法对小波包频带系数进行重构,得到以各频带信号样本熵值为元素的反映机组故障信息的特征向量, 最后采用LVQ神经网络对试验振动信号进行分类,验证结果表明:基于小波包变换与样本熵相结合的特征提取方法对水泵机组不同振动状态具有较好的区分度,是一种合适的水泵机组故障特征提取方法。

Abstract

Most large-scale pump units have installed condition monitoring systems currently. How to extract fault features from the original data is a major and hot focus of the water pump faults diagnosis. The paper based on wavelet analysis, presents a method combining wavelet packet transform with sample entropy for signal feature extraction. Firstly, it decomposes the vibration signal by wavelet packet transform to obtain wavelet packet coefficients. Then, it reconstructs wavelet packet coefficients through the method of wavelet packet transform combining samples entropy algorithm to obtain feature vectors consisting of each band signal sample entropy characteristic elements. Finally, the method combining wavelet packet transform with sample entropy is proved, by the analysis of the vibration signal along with LVQ neural network, to play a good performance in identifying different running condition of water pump units. It is a suitable method to extract fault features from pump units.

关键词

水泵机组 / 小波分析 / 小波包分析 / 样本熵

Key words

pump unit / wavelet analysis / wavelet packet analysis / sample entropy

基金

水利部"948"项目"大型泵机组全方位在线监测与诊断系 统 "(201321 ); 
国家自然科学基金资 助项目(51379160)。

引用本文

导出引用
宋媛 彭利鸿 赖冠文 张嘉勋 肖志怀 宋丽波. 基于小波包和样本熵的水泵机组振动特征提取[J].中国农村水利水电, 2017(3): 146-152
SONG Yuan, PENG Li-hong, LAIGuan-wen, ZHANG Jia-xun, XIAO Zhi-huai, SONG Li-bo. Vibration Feature Extraction of Pump Unit Based on Wavelet Packet and Sample Entropy[J].China Rural Water and Hydropower, 2017(3): 146-152

参考文献

[1] 刘晓亭,冯辅周. 水电机组运行设备监测诊断技术及应用[M].北京: 中国水利水电出版社,2010.
[2] 陈喜阳.水电机组状态检修中若干关键技术研究[D].武汉: 华中科技大学,2005.
[3] Richman J S,Moorman J R. Physiological time-series analysis usingapproximate entropy and sample entropy [J]. American Journal ofPhysiology-Heart and Circulatory Physiology,2002,278( 6) : 2 039-2 049.
[4] R Alcaraz,J J Rieta. A review on sample entropy applications for thenon-invasive analysis of atrial fibrillation electrocardiograms[J]. Biomedical Signal Processing and Control,2010.
[5] Raffa F A,Vatta V. The dynamic stiffness method for linear rotor-bearing system[J]. ASME Journal of Vibration and Acoustics,1996,118( 7) : 332-339.
[6] Pincus S M. Assessing serial irregularity and its implications for health[J]. Ann. N. Y. Acad. Sci,2002,954: 245-267.
[7] 平 源. 基于支持向量机的聚类及文本分类研究[D]. 北京: 北京邮电大学 2012.
[8] 彭 璐. 支持向量机分类算法研究与应用[D]. 长沙: 湖南大学2007.
[9] 瞿 曌,赖 旭,盛 旺.水电机组状态检修模式探讨[J].华东电力,2008,( 1) : 109-111.
[10] 王洪杰,刘全忠,季天晶,等. 锅炉给水泵故障机理及诊断技术的研究[J]. 节能技术,2004,22( 6) : 30-31.
[11] 丁世飞,齐丙娟,谭红艳. 支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40( 1) : 2-10.
[12] 吴兴伟,王 雷,迟道才. 基于多分类 SVM 的水泵故障诊断的研究[J]. 节水灌溉,2009,( 7) : 37-39.
[13] 张 炜,张 磊,李 亮. 基于 GA 优化的 SVM 涡轮泵故障诊断[J].液压与气动,2009,( 1) : 84-86.
[14] 苟 博,黄贤武. 支持向量机多类分类方法[J]. 数据采集与处理,2006,21( 3) : 334-339.
PDF(2553 KB)

访问

引用

详细情况

段落导航
相关文章

/