基于小波多尺度混沌特征参数的离心泵汽蚀故障诊断

PDF(1367 KB)

振动与冲击 ›› 2021, Vol. 40 ›› Issue (21) : 106-112.

论文

梁超，周云龙，杨宁

作者信息 +

Cavitation fault diagnosis of centrifugal pump based on wavelet multi-scale chaotic characteristic parameters

LIANG Chao, ZHOU Yunlong, YANG Ning

Author information +

文章历史 +

摘要

提出了基于小波多尺度混沌特征参数-最小二乘支持向量机（LS-SVM）的故障诊断模型。首先依据实验现象和入口压力信号的分频带混沌特征参数，将不同的有效汽蚀余量（NPSHa）划分为4种状态；然后对入口压力信号进行3个尺度的小波分解，构建出各尺度低频信号的混沌参数特征向量；最后应用经遗传算法优化的LS-SVM实现汽蚀故障诊断。结果表明：利用分频带混沌特征能够准确描述汽蚀状态及其演变规律，更符合离心泵的实际运行情况。模型的故障诊断精度在87.5%以上，可以高效识别出离心泵的不同汽蚀状态。

Abstract

A fault diagnosis model based on wavelet multi-scale chaotic characteristic parameters-least square support vector machine (LS-SVM) was proposed. Firstly, the different net positive suction head available (NPSHa) values were classified into 4 states according to the experimental phenomena and the chaotic characteristic parameters of frequency bands of the inlet pressure signals. Then, the pressure signals were decomposed by wavelet at 3 scales, and the chaotic parameter feature vectors of low-frequency signals at each scale were constructed. Finally, the LS-SVM which was optimized by genetic algorithm was used to realize cavitation fault diagnosis. Results show that the cavitation states and their evolution laws can be accurately described by using the chaotic characteristics of frequency bands, which is more in line with the actual operation condition of the centrifugal pump. The fault diagnosis accuracy of the model was above 87.5%, which can identify the different cavitation states in the centrifugal pump efficiently.

导出引用

梁超，周云龙，杨宁. 基于小波多尺度混沌特征参数的离心泵汽蚀故障诊断[J]. 振动与冲击, 2021, 40(21): 106-112

LIANG Chao, ZHOU Yunlong, YANG Ning. Cavitation fault diagnosis of centrifugal pump based on wavelet multi-scale chaotic characteristic parameters[J]. Journal of Vibration and Shock, 2021, 40(21): 106-112

参考文献

[1] 查森.叶片泵原理及水力设计[M].北京:机械工业出版社，1988.
Zha Sen. Principle and Hydraulic Design of Vane Pump[M]. Beijing: China Machine Press，1988.
[2] 周云龙，梁超.基于EMD能量熵的离心泵汽蚀故障诊断研究方法[J].化工自动化仪表，2010，37（5）:41-46.
ZHOU Yunlong，LIANG Chao. Research on fault dignosis of vavitation for a centrifugal pump based on energy of EMD[J].Contral and Instruments in Chemical Industry，2010，37(5):41-46.
[3] 周云龙，刘永奇，薛广鑫.基于EMD和边际谱带能量的离心泵汽蚀故障诊断[J].化工学报，2012，63（2）:545-550.
ZHOU Yunlong，LIU Yongqi，XUE Guangxin.Fault diagnosis of cavitation forcentrifugal pump based on EMD and HHT marginal spectrumenergy[J]．Journal of Chemical Industry and Engineering，2012，63(2):545-550．
[4] 周云龙，吕远征.基于多点噪声分析的离心泵早期汽蚀故障诊断[J].振动与冲击，2017，36（7）:39-44.
ZHOU Yunlong，LÜ Yuanzheng.Incipient cavitation fault diagnosis for a centrifugal pump based on multi-position noise      analysis[J].Journal of Vibration and Shock，2017，36(7):39-44.
[5] 郝点，周昌静，张志飞.相关小波分析用于离心泵初生汽蚀诊断[J].石油大学学报(自然科学版)，2005，29(2):78-82.
HAO Dian，ZHOU Changjing，ZHANG Zhifei. Diagnosis of incipient cavitation in centrifugal pump using wavlet combined with autocorrelation methord[J]. Journal of the University of Petroleum(Edition of Natural Science)，2005，29(2):78-82.
[6] 卿红，韩华，崔晓钰等.基于粒子群优化最小二乘支持向量机的离心式制冷机故障诊断[J].暖通空调，2018,48(9)：120-126.
QING Hong，HAN Hua，CUI Xiaoyu et al. Fault diagnosis of centrifugal chillers based on particle swarm optimization- least square support vector machine[J]. Journal of Heating Ventilating and Airconditioning, 2018,48(9)：120-126.
[7] 胡良谋，曹克强，苏新兵等.基于LS-SVM的飞控系统传感器故障诊断[J].飞行力学,2011，29(3):36-39.
    HU Liangmou，CAO Keqiang，SU Xinbing et al. Sensor Fault diagnosis of Flight Control System based on LS-SVM[J].Flight Dynamics，2011，29(3):36-39.
[8] 赵孝礼,赵荣珍.全局与局部判别信息融合的转子故障数据集降维方法研究[J].自动化学报, 2017，43(4): 560 -567.
ZHAO Xiaoli, ZHAO Rongzhen. A method of dimension reduction of rotor faults data set based on fusion of global and local discrimi-nant information[J]. Acta Automatica Sinica，2017，43(4): 560-567.
[9] 苏祖强，萧红，张毅等.基于小波包分解与主流形识别的非线性降噪[J].仪器仪表学报，2016，37（9）:1954-1961.
SU Zuqiang，XIAO Hong，ZHANG Yi et al.Nonlinear noise reduction method based on wavelet packet decomposition and principle mainfold learning[J].Chinese Journal of Science Instrument，2016，37(9):1954-1961.
[10] 何正嘉，訾艳阳，张西宁.现代信号处理及工程应用[M].西安：西安交通大学出版社，2007.
HE Zhengjia，ZI Yanyang，ZHANG Xining.Modern signal processing and engineering applications[M].Xi’an: Xi’an Jiaotong University Press，2007.
[11] Donald B P，Andrew T W.Wavelet method for time series analysis[M]. Beijing: China Machine Press，2004.
[12] 吕金虎.混沌时间序列分析及其应用[M].武汉：武汉大学出版社，2002.
LÜ Jinhu.Chatics time series analysis and application[M]. Wuhan: Wuhan University Press，2002.
[   [13] Rosenstein M T，Collins J，Deluca C J.A practical method for calculating largest Lyapunov exponents from small data sets[J]. Physica D，1993，65(1-2)：117-134.
[14] 张文超.摆动条件下自然循环流动不稳定性混沌特性分析及预测[D]. 哈尔滨：哈尔滨工程大学，2013.
ZHANG Wenchao.Chaotic analysis and prediction on flow instability of natural circulation under rolling motion condition[D].Harbin:Harbin Engineering University，2013.
[15] 余波，李应红，张朴.关联维数和Kolmogorov熵在航空发
动机故障诊断中的应用[J].航空动力学报，2006，21（1）：219-224.
YU Bo，LI Yinghong，ZHANG Pu.Applacation of correlation dimension and Kolmogorov entropy in aeroengine fault diagnosis[J]. Journal of aerospace power，2006，21（1）：219-224.
[16] 孙斌，刘天栋，周云龙.小波包主成分分析在气液两相流流型识别中的应用[J].吉林大学学报（工学版），2009，39(6):1532-1537.
SUN Bin，LIU Tiandong，ZHOU Yunlong. Application of principal component analysis of wavelet packet on flow   regime identification of gas-liquid two-phase flow[J].Journal of Jilin University(Engineering ang Technology Edition)，2009，39(6)：1532-1537.
[17] 孙一睿，李钰鑫，陈磊等.基于遗传算法优化神经网络的SCR催化剂失效预测[J].中国电机工程学报，2016，36(9):112-120.
Sun Y R，Li Y X，Chen L，et al.Deactivation forecasting of SCR catalyst based on GAMP neural networks[J]. Proceedings of the CSEE，2016，36(9):112-120.
[18] Alencar A S C，Rochaneto A R，Gomes J P P.A new pruning method for extreme learning machines via genetic algorithms
     [J].Applied Soft Computing Journal，2016，44:101-107.
[19] 徐家祖，尹勐.离心泵的低频汽蚀脉动现象[J].流体机械，1998，26(1):5-9.
Xu J Z，Yin M. Low frequency fluctuation phenomenon of cavitation in a centrifugal pump[J].Fluid Mechinary，1998，26(1):5-9.