Please wait a minute...
工程设计学报  2024, Vol. 31 Issue (6): 741-749    DOI: 10.3785/j.issn.1006-754X.2024.03.410
【特约专栏】“双碳”背景下新型能源装备设计、制造、运维关键技术及其应用     
基于小波包分解与随机森林的离心泵故障诊断
马飞1(),邵礼光1,徐君1,陶梦秋1,袁沛1(),胡炳涛2
1.杭州景业智能科技股份有限公司,浙江 杭州 310053
2.浙江大学 机械工程学院,浙江 杭州 310028
Centrifugal pump fault diagnosis based on wavelet pack decomposition and random forest
Fei MA1(),Liguang SHAO1,Jun XU1,Mengqiu TAO1,Pei YUAN1(),Bingtao HU2
1.Hangzhou Boomy Intelligent Technology Co. , Ltd. , Hangzhou 310053, China
2.School of Mechanical Engineering, Zhejiang University, Hangzhou 310028, China
 全文: PDF(4869 KB)   HTML
摘要:

针对核电厂离心泵在线故障诊断困难的问题,提出了一种基于小波包分解与随机森林的故障诊断方法。首先,利用小波包分解对离心泵电机驱动端径向垂直方向的振动信号进行3层分解并提取子频带能量特征。然后,基于离心泵振动信号的波形数据提取时域统计特征,并与小波包能量特征相结合作为随机森林模型的输入。最后,通过由振动试验得到的离心泵振动数据集对随机森林模型进行训练,形成离心泵故障诊断模型,并对该模型与支持向量机、逻辑斯蒂回归、K近邻、高斯朴素贝叶斯等机器学习模型在相同的离心泵振动数据集上进行了对比测试。结果表明,所构建的模型能够准确识别离心泵正常、叶轮破损、叶轮堵塞、电机轴承故障等运行状态,并表现出更优的分类性能。基于小波包分解与随机森林的故障诊断方法可以有效地从振动信号中提取特征并实现故障分类,对于核电厂离心泵在线故障智能诊断具有一定的可行性和有效性。

关键词: 离心泵故障诊断振动信号小波包分解随机森林    
Abstract:

Aiming at the difficulties of on-line fault diagnosis of centrifugal pumps in nuclear power plants, a fault diagnosis method based on wavelet pack decomposition and random forest is proposed. Firstly, the wavelet pack decomposition was used to decompose the vibration signal in the radial vertical direction of the centrifugal pump motor drive end into three layers, and the sub-band energy features were extracted. Then, the time-domain statistical features were extracted based on the waveform data of centrifugal pump vibration signal, and combined with wavelet packet energy features as inputs for the random forest model. Finally, the random forest model was trained with centrifugal pump vibration dataset collected from vibration test, and the centrifugal pump fault diagnosis model was formed. This model was compared with machine learning models such as support vector machine, logistic regression, K-nearest neighbor and Gaussian Naive Bayes on the same centrifugal pump vibration dataset. The results showed that the constructed model could accurately identify different operating states of the centrifugal pump, such as normal operation, impeller damage, impeller blockage and motor bearing fault, and exhibited better classification performance. The fault diagnosis method based on wavelet packet decomposition and random forest can effectively extract features from vibration signals and realize fault classification, which has certain feasibility and effectiveness for on-line fault intelligent diagnosis of centrifugal pumps in nuclear power plants.

Key words: centrifugal pump    fault diagnosis    vibration signal    wavelet pack decomposition    random forest
收稿日期: 2023-12-20 出版日期: 2024-12-31
CLC:  TH 311  
基金资助: 2022年度浙江省“尖兵”“领雁”研发攻关计划项目(2022C01054)
通讯作者: 袁沛     E-mail: maf@boomy.cn;yuanp@boomy.cn
作者简介: 马 飞(1991—),男,工程师,硕士,从事设备故障诊断研究,E-mail: maf@boomy.cn,https://orcid.org/0009-0000-7331-114X
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
马飞
邵礼光
徐君
陶梦秋
袁沛
胡炳涛

引用本文:

马飞,邵礼光,徐君,陶梦秋,袁沛,胡炳涛. 基于小波包分解与随机森林的离心泵故障诊断[J]. 工程设计学报, 2024, 31(6): 741-749.

Fei MA,Liguang SHAO,Jun XU,Mengqiu TAO,Pei YUAN,Bingtao HU. Centrifugal pump fault diagnosis based on wavelet pack decomposition and random forest[J]. Chinese Journal of Engineering Design, 2024, 31(6): 741-749.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn.1006-754X.2024.03.410        https://www.zjujournals.com/gcsjxb/CN/Y2024/V31/I6/741

图1  离心泵故障诊断系统框架
特征量计算式
有效值xrms=1Mm=1Mxm2
均值xˉ=1Mm=1Mxm
标准差σx=1M-1m=1Mxm-xˉ2
峰值因子Cf=maxxmxrms
波形因子Sf=xrms1Mm=1Mxm
峭度Kr=m=1Mxm-xˉ4Mσx4
偏度Sk=m=1Mxm-xˉ3Mσx3
扰动因子Pf=σxxˉ
表1  振动信号时域统计特征
图2  三层小波包分解结构
图3  随机森林模型构建流程
参数数值
供电电压/V220~240
电机功率/W373
流量/(m3/h)5.796
叶轮直径/mm118.88
叶片数量/片3
表2  离心泵规格参数
图4  离心泵振动试验台示意图
图5  离心泵振动试验台实物图
图6  试验用离心泵的缺陷部件
图7  离心泵的原始振动信号
图8  正常状态下离心泵振动信号的小波包分解系数
图9  正常状态下离心泵的原始振动信号及重构信号的时域波形和频谱图
图10  离心泵振动信号的小波包能量归一化值
图11  基于随机森林的离心泵故障诊断结果
图12  不同故障诊断模型的预测结果
模型准确率/%AUC-ROC召回率F1
SVM98.00.999 30.980 00.980 3
KNN99.51.000 00.995 00.995 0
随机森林100.01.000 01.000 01.000 0
逻辑斯蒂回归99.50.997 20.995 00.995 0
高斯朴素贝叶斯98.50.997 10.985 00.985 0
表3  不同故障诊断模型的预测效果对比
1 孔令杰, 雒晓辉, 宋立, 等. 核电厂离心泵叶轮裂纹故障分析与研究[J]. 水泵技术, 2019(4): 37-40.
KONG L J, LUO X H, SONG L, et al. Analysis and study of the impeller crack fault diagnosis for nuclear power plant centrifugal pump[J]. Pump Technology, 2019(4): 37-40.
2 黄若琳, 张志明, 李栋梁. 核电厂卧式多级离心泵智能诊断系统方案设计[J]. 新型工业化, 2022, 12(11): 104-107.
HUANG R L, ZHANG Z M, LI D L. Scheme design of intelligent diagnosis system for horizontal multistage centrifugal pump in nuclear power plant[J]. The Journal of New Industrialization, 2022, 12(11): 104-107.
3 宋怡. 基于贝叶斯分类的核电站泵类设备振动故障诊断方法[J]. 机械制造与自动化, 2023, 52(4): 64-67.
SONG Y. Vibration fault diagnosis method of pump equipment in nuclear power plant based on Bayesian classification[J]. Machine Building & Automation, 2023, 52(4): 64-67.
4 ABOSHOSHA A, HAMAD H A. Employing adaptive fuzzy computing for RCP intelligent control and fault diagnosis[J]. Nuclear Science and Techniques, 2023, 34: 138.
5 DUTTA N, KALIANNAN P, SHANMUGAM P. SVM algorithm for vibration fault diagnosis in centrifugal pump[J]. Intelligent Automation & Soft Computing, 2023, 35(3): 2997-3020.
6 辜文娟, 张扬. 基于IMIE、MCFS和SSA-ELM的离心泵故障诊断方法[J]. 机电工程, 2023, 40(9): 1456-1463.
GU W J, ZHANG Y. Fault diagnosis method of centrifugal pump based on IMIE and SSA-ELM[J]. Journal of Mechanical & Electrical Engineering, 2023, 40(9): 1456-1463.
7 KARAGIOVANIDIS M, PANTAZI X E, PAPAMICHAIL D, et al. Early detection of cavitation in centrifugal pumps using low-cost vibration and sound sensors[J]. Agriculture, 2023, 13(8): 1544.
8 ARASTE Z, SADIGHI A, JAMIMOGHADDAM M. Fault diagnosis of a centrifugal pump using electrical signature analysis and support vector machine[J]. Journal of Vibration Engineering & Technologies, 2023, 11(5): 2057-2067.
9 王治国, 吴茜. 基于共振特征的离心泵故障诊断[J]. 自动化应用, 2023(7): 144-146.
WANG Z G, WU Q. Fault diagnosis of centrifugal pump based on resonance characteristics[J]. Automation Application, 2023(7): 144-146.
10 CHEN Y F, YUAN J P, LUO Y, et al. Fault prediction of centrifugal pump based on improved KNN[J]. Shock and Vibration, 2021, 2021(1): 7306131.
11 TOBI M AL, BEVAN G, WALLACE P, et al. Faults diagnosis of a centrifugal pump using multilayer perceptron genetic algorithm back propagation and support vector machine with discrete wavelet transform-based feature extraction[J]. Computational Intelligence, 2021, 37(1): 21-46.
12 MANIKANDAN S, DURAIVELU K. Vibration-based fault diagnosis of broken impeller and mechanical seal failure in industrial mono-block centrifugal pumps using deep convolutional neural network[J]. Journal of Vibration Engineering & Technologies, 2023, 11(1): 141-152.
13 孙原理, 宋志浩. 基于多物理场信号相关分析与支持向量机的离心泵故障诊断方法[J]. 振动与冲击, 2022, 41(6): 206-212.
SUN Y L, SONG Z H. Centrifugal pump fault diagnosis based on the multi-physical field signals correlation analysis and support vector machine[J]. Journal of Vibration and Shock, 2022, 41(6): 206-212.
14 RAPUR J S, TIWARI R. Experimental time-domain vibration-based fault diagnosis of centrifugal pumps using support vector machine[J]. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, 2017, 3(4): 044501.
15 杨波, 黄倩, 付强, 等. 基于CEEMD和优化KNN的离心泵故障诊断方法[J]. 机电工程, 2022, 39(11): 1502-1509.
YANG B, HUANG Q, FU Q, et al. Fault diagnosis method for horizontal centrifugal pump based on CEEMD and optimized KNN[J]. Journal of Mechanical & Electrical Engineering, 2022, 39(11): 1502-1509.
16 AHMAD Z, HASAN M J, KIM J M. Centrifugal pump fault diagnosis using discriminative factor-based features selection and K-nearest neighbors[C]//International Conference on Intelligent Systems Design and Applications. Cham: Springer International Publishing, 2022: 145-153.
17 MURALIDHARAN V, SUGUMARAN V. Feature extraction using wavelets and classification through decision tree algorithm for fault diagnosis of mono-block centrifugal pump[J]. Measurement, 2013, 46(1): 353-359.
18 ADEODU A, DANIYAN I, OMITOLA O, et al. An adaptive Industrial Internet of things (IIOts) based technology for prediction and control of cavitation in centrifugal pumps[J]. Procedia CIRP, 2020, 91: 927-934.
19 张胜文, 杨凌翮, 程德俊. 数字孪生驱动的离心泵机组故障诊断方法研究[J]. 计算机集成制造系统, 2023, 29(5): 1462-1470.
ZHANG S W, YANG L H, CHENG D J. Fault diagnosis method of centrifugal pump driven by digital twin[J]. Computer Integrated Manufacturing Systems, 2023, 29(5): 1462-1470.
20 KUMAR A, TANG H S, VASHISHTHA G, et al. Noise subtraction and marginal enhanced square envelope spectrum (MESES) for the identification of bearing defects in centrifugal and axial pump[J]. Mechanical Systems and Signal Processing, 2022, 165: 108366.
[1] 谢苗,孟庆爽,李博,卢进南,李玉岐,杨志勇. 融合短时傅里叶变换和卷积神经网络的托辊故障诊断方法[J]. 工程设计学报, 2024, 31(5): 565-574.
[2] 叶子汉,王中华,姜潮,吕新,张哲. 基于多判别器辅助分类器生成对抗网络的故障诊断方法研究[J]. 工程设计学报, 2024, 31(2): 137-150.
[3] 纪超,王亮,王孝敬,李小兵,曹雯. 基于MSSA-SVM的电缆隧道故障预警系统设计[J]. 工程设计学报, 2023, 30(1): 109-116.
[4] 崔旭浩,郗欣甫,孙以泽. 数据驱动的经编机横移机构故障检测方法研究[J]. 工程设计学报, 2022, 29(3): 263-271.
[5] 吴国沛, 余银犬, 涂文兵. 永磁同步电机故障诊断研究综述[J]. 工程设计学报, 2021, 28(5): 548-558.
[6] 倪洪启, 金驰, 冯霏. 波纹补偿器故障诊断系统研制[J]. 工程设计学报, 2019, 26(3): 354-363.
[7] 马天兵, 王孝东, 杜菲, 王鑫泉. 基于GA-SVM的刚性罐道故障诊断[J]. 工程设计学报, 2019, 26(2): 170-176.
[8] 胡艺耀, 朱斌, 张伟, 何畏, 沈平生. 知识库构建工具软件的设计与实现[J]. 工程设计学报, 2018, 25(4): 367-373.
[9] 张强, 刘志恒, 王海舰, 张赫哲. 截齿磨损程度的多特征信号融合识别研究[J]. 工程设计学报, 2018, 25(3): 278-287.
[10] 李晓豁,翁正洋,钱亚森,史尚伟,李 岩. 基于果蝇算法优化模糊RBF网络的液压破碎锤故障诊断[J]. 工程设计学报, 2015, 22(6): 540-545.
[11] 李玲玲, 景丽婷, 马东娟, 李志刚. 改进证据理论及其在电力系统故障诊断中的应用[J]. 工程设计学报, 2012, 19(6): 485-488.
[12] 韩海涛,马红光,韩琨,郑耿乐. 关于Volterra频域核辨识的多音激励信号设计[J]. 工程设计学报, 2012, 19(2): 123-127.
[13] 郝志勇, 刘伟, 夏玮, 闫闯. 基于BP神经网络的吸运风机故障诊断[J]. 工程设计学报, 2012, 19(1): 57-60.
[14] 韩海涛, 马红光, 李飞, 张家良. 基于系统输出频率响应函数的非线性系统研究[J]. 工程设计学报, 2011, 18(5): 373-376.
[15] 郭玉秀, 倪晓红, 王玉田. 基于混沌弱信号检测的轧机故障诊断研究[J]. 工程设计学报, 2011, 18(3): 218-221.