基于CEEMD-LSTM的离心泵偏工况诊断方法研究

摘要
图/表
参考文献(15)
相关文章 (2)

全文: PDF (2476 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要离心泵在各行业中应用十分广泛，耗电量巨大。离心泵偏工况运行时，内部流动会趋于紊乱，导致效率下降，能耗上升。本文针对离心泵偏工况振动信号变化微弱和强干扰的特点，采用双通道信息融合，利用互补集合经验模态分解，对振动信号进行时序特征提取，结合长短时记忆模型智能识别，构建离心泵偏工况诊断模型。仿真信号对比不同预处理方法，凸显了互补集合经验模态分解模型的特征提取能力；验证了工况状态与低频振动信号的相关性，经过实验数据对比分析，进一步验证了模型优越性，测试准确率达98.5%。该方法可以监测离心泵运行工况，保证运行效率。
关键词：偏工况；互补集合经验模态；长短时记忆模型

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	刘荣伟1
	何伟挺2
	汪琳琳1
	杨帅1
	武鹏1
	吴大转1

关键词 ：偏工况, 互补集合经验模态, 长短时记忆模型

Abstract：Centrifugal pumps are widely used in various industries and consume power heavily. When centrifugal pump operates under partial working conditions, the internal flow tends to be chaotic, resulting in a decline in efficiency and an increase in energy consumption. In this paper, according to weak variation and strong interference of vibration signals in off-working condition of centrifugal pump, two-channel information fusion and complementary set empirical mode decomposition(CEEMD) were adopted to extract the time-series features of vibration signals. Combined with the intelligent recognition of long-short time memory(LSTM) model, a diagnostic model was established for off-working condition of centrifugal pump. The simulation signals are compared with different preprocessing methods to highlight the feature extraction ability of CEEMD. The correlation is verified between working conditions and low-frequency vibration signals. The superiority of the model is further verified through comparative analysis of experimental data, and the test accuracy rate reaches 98.5%. This method can monitor the running condition of centrifugal pump and ensure the running efficiency.
Key words: Partial conditions; Empirical mode of complementary set; Long and short time memory model

Key words： Partial conditions Empirical mode of complementary set Long and short time memory model

收稿日期: 2021-05-19 出版日期: 2022-10-15

引用本文:

刘荣伟1，何伟挺2，汪琳琳1，杨帅1，武鹏1，吴大转1. 基于CEEMD-LSTM的离心泵偏工况诊断方法研究[J]. 振动与冲击, 2022, 41(19): 114-121.
LIU Rongwei1, HE Weiting2, WANG Linlin1, YANG Shuai1, WU Peng1, WU Dazhuan1. CEEMD-LSTM-based diagnosis method for off-design working conditions of centrifugal pump. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(19): 114-121.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2022/V41/I19/114

[1] 何玉芬, 张镭漓, 王振龙. 国内离心泵节能措施浅析[J]. 水利水电快报. 2019, 40(04): 49-52.
HE Yufang, ZHANG Leili, WANG Zhenlong. Brief analysis of domestic centrifugal pump energy saving measures[J]. Hydroelectricity Bulletin. 2019, 40(04): 49-52.
[2] SUN H, Yuan S Q, Luo Y, et al. Unsteady characteristics analysis of centrifugal pump operation based on motor stator current[J]. Proceedings of the Institution of Mechanical Engineers, 2017, 231(8): 689-705.
[3] 韩岳江. 基于无传感器技术的离心泵工况诊断系统研发[D]. 镇江: 江苏大学, 2020.
HAN Yuejiang, Research and development of centrifugal pump working condition diagnosis system based on sensorless technology[D]. Zhenjiang: Jiangsu University, 2020.
[4] 陈长盛, 马俊, 柳瑞锋, 等. 运行工况对离心泵振动影响的试验研究[J]. 噪声与振动控制. 2012, 32(06): 199-202.
CHEN Changsheng, MA Jun, LIU Ruifeng, et al. Experimental Study on Centrifugal Pump Vibration in Different Operation Conditions[J]. Noise and Vibration Control. 2012, 32(06): 199-202.
[5] 周林玉. 偏离工况下离心泵的压力脉动和振动分析[J]. 流体机械, 2015, 43(2): 52-55.
ZHOU Linyu. Analysis on Pressure Fluctuation and Vibration of a Centrifugal Pump for Off-design Conditions[J]. Fluid Machinery, 2015, 43(2): 52-55.
[6] GUZZOMI A, PAN J. Monitoring single-stage double-suction pump efficiency using vibration indicators[J]. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering. 2014, 228(4): 332-336.
[7] ZHANG M, JIANG ZN, FENG K. Research on variational mode decomposition in rolling bearings fault diagnosis of the multistage centrifugal pump[J]. Mechanical Systems and Signal Processing. 2017, 93: 460-493.
[8] WANG H, LI S, SONG L, et al. A novel convolutional neural network based fault recognition method via image fusion of multi-vibration-signals[J]. Computers in Industry. 2019, 105: 92-104.
[9] 伍柯霖, 钱全, 邢允, 等. 基于时频分析的离心泵空化状态表征研究[J]. 工程热物理学报. 2021, 42(01):106-114.
WU Kelin, QIAN Quan, XING Yun, et al. Research on cavitation characterization of centrifugal pumps based on timefrequency analysis[J]. Journal of Engineering Thermophysics. 2021, 42(01):106-114.
[10] 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(07): 39-44.
ZHOU Yunlong, LV Yuanzheng. Incipient cavitations fault diagnosis for a centrifugal pump based on multi-position noise analysis[J]. Journal of Vibration and Shock. 2017, 36(07):39-44.
[11] 龚波. 基于MCSA的离心泵运行工况监测基础研究[D]. 镇江: 江苏大学, 2019.
GONG Bo. Fundamental Research on Monitoring of Operating Conditions for Centrifugal Pump Based on MCSA[D]. Zhenjiang: Jiangsu University, 2019.
[12] HUANG N E. The empirical mode decomposition method and the Hilbert spectrum for non-stationary time series analysis[J]. The Royal Society. 1998, 454(1971): 903-995.
[13] ZHENG Y. Computer-assisted diagnosis for chronic heart failure by the analysis of their cardiac reserve and heart sound characteristics[J]. Computer Methods & Programs in Biomedicine. 2015, 122(3): 372-383.
[14] 董利超, 郭兴明, 郑伊能. 基于CEEMD 的心音信号小波包去噪算法研究[J]. 振动与冲击. 2019, 38(9): 192-221.
DONG Lichao, GUO Xingming, ZHENG Yineng. Wavelet packet de-noising algorithm for heart sound signals based on CEEMD[J]. Journal of Vibration and Shock. 2019, 38(9): 192-221.
[15] 宋永兴. 基于主成分分析的水力旋转机械低频声特征提取方法研究[D]. 杭州: 浙江大学, 2019.
SONG Yongxing. Study on low frequency acoustic feature extraction method of hydraulic rotating machinery based on principal component analysis[D]. Zhejiang University, 2019