基于AWTD和LSTM的循环泵口环磨损趋势预测研究Research on the wear trend prediction of circulating pump mouth ring based on AWTD and LSTM
魏道红,于雅静,林彬,刘瀚文,于涵
Wei Daohong,Yu Yajing,Lin Bin,Liu Hanwen,Yu Han
摘要(Abstract):
因受流体噪声与多源机械干扰的叠加影响,循环泵口环磨损状态的早期识别与趋势预测面临挑战。针对这一问题,提出一种基于自适应小波阈值降噪(AWTD)和长短期记忆神经网络(LSTM)的循环泵口环磨损趋势预测方法。通过搭建循环泵试验测试平台,分别采集口环间隙为0.2,0.5,1.0 mm三种不同磨损程度的振动信号,并利用连续小波变换(CWT)进行多尺度时频特征分解;利用基于信号能量分布的自适应阈值函数,对含噪振动信号实施降噪处理,降噪后信号的信噪比R_(SN)分别提升0.32dB,18.34dB和15.30dB,有效凸显磨损特征成分,进一步构建LSTM预测模型。结果表明,模型的均方根误差E_(RMS)低至0.021 039,平均绝对误差E_(MA)为0.016 598,说明模型具有较高的预测精度,进一步验证该方法在循环泵口环磨损早期趋势预测中的有效性。
Due to the combined influence of fluid noise and multiple mechanical disturbances,the early identification and trend prediction of wear conditions of circulating pump mouth ring face challenges.To address this issue,a method for predicting the wear trend of circulating pump was proposed based on adaptive wavelet threshold denoising(AWTD) and long short-term memory(LSTM).By establishing circulating pump experimental test platform,the vibration signals were collected under three different wear levels(mouth ring clearance of 0.2,0.5,1.0 mm),and continuous wavelet transforms(CWT) was used to perform multi-scale time/frequency feature decomposition.By using an adaptive threshold function based on signal energy distribution,noise reduction processing was carried out on the noisy vibration signals.After noise reduction,the signal-to-noise ratio R_(SN) of the signals increased by 0.32dB,18.34dB and 15.30dB respectively,effectively highlighting the wear feature components,and further constructing an LSTM prediction model.The results show that the root mean square error E_(RMS) of the model is as low as 0.021 039,and the mean absolute error E_(MA) is 0.016 598,indicating that the model has high prediction accuracy and verifies the effectiveness of this method in early trend prediction of wear in circulating pump mouth ring.
关键词(KeyWords):
循环泵;口环磨损;自适应小波阈值降噪(AWTD);长短期记忆神经网络(LSTM);趋势预测
circulating pump;wear of mouth ring;adaptive wavelet threshold denoising(AWTD);long short-term memory(LSTM);trend prediction
基金项目(Foundation): 国家自然科学基金青年科学基金项目(52409113)
作者(Author):
魏道红,于雅静,林彬,刘瀚文,于涵
Wei Daohong,Yu Yajing,Lin Bin,Liu Hanwen,Yu Han
参考文献(References):
- [1]周昌静,陈战,庄法坤,等.口环磨损对离心泵性能的影响及故障诊断研究[J].石油化工设备,2020,49(2):6-9.Zhou C J,Chen Z,Zhuang F K,et al.Study on the influence of ring wear on centrifugal pump performance and fault diagnosis[J].PetroChemical Equipment,2020,49(2):6-9.
- [2]耿要强,姚睿,付兵非,等.阀冷主泵状态监测与故障诊断技术研究[J].自动化仪表,2024,45(6):99-103.Geng Y Q,Yao R,Fu B F,et al.Study of condition monitoring and fault diagnosis technology of valve cooled main pump[J].Process Automation Instrumentation,2024,45(6):99-103.
- [3]Qian H,Zhi Y F,Zhang R Y.Design of intelligent diagnosis operation and maintenance system for circulating water pump[C]//2021 Global Reliability and Prognostics and Health Management(PHM-Nanjing).IEEE,2021:1-4.
- [4]董庆兵,汪汉明,冯成程,等.核电循环泵行星齿轮热弹流润滑研究[J].润滑与密封,2024,49(2):1-10.Dong Q B,Wang H M,Feng C C,et al.Study on thermal elastohydrodynamic lubrication of planetary gears in circulating water pump of nuclear power plant[J].Lubrication Engineering,2024,49(2):1-10.
- [5]刘朝玺,叶志锋,王彬,等.基于人工神经网络的离心泵空化故障诊断[J].机械制造与自动化,2024,53(4):151-153.Liu C X,Ye Z F,Wang B,et al.Centrifugal pump cavitation fault diagnosis based on artificial neural network[J].Machine Building&Automation,2024,53(4):151-153.
- [6]徐克凡,陈佳惠,蒋华,等.水利工程泵站机电设备故障诊断方法及检修重要性[J].流体测量与控制,2024,5(5):11-14.Xu K F,Chen J H,Jiang H,et al.Fault diagnosis method and maintenance importance of mechanical and electrical equipment in pumping station of hydraulic engineering[J].Fluid Measurement&Control,2024,5(5):11-14.
- [7]陈公朴.大型水泵机组振动趋势预测方法研究[J].工程技术研究,2024,9(7):38-40.Chen G P.Research on vibration trend prediction method of large water pump unit[J].Engineering Technology Research,2024,9(7):38-40.
- [8]Gao J Y,Ma L Y,Qing C,et al.A health monitoring model for circulation water pumps in a nuclear power plant based on graph neural network observer[J].Sensors,2024,24(14):4486.
- [9]Sanayha M,Vateekul P.Fault detection for circulating water pump using time series forecasting and outlier detection[C]//2017 9th International Conference on Knowledge and Smart Technology (KST).IEEE,2017:193-198.
- [10]Wang J J,Zhang L B,Zheng Y H,et al.Adaptive prognosis of centrifugal pump under variable operating conditions[J].Mechanical Systems and Signal Processing,2019,131:576-591.
- [11]汪家琼,胡波,黄倩,等.离心泵轴承疲劳寿命仿真分析及试验验证[J].排灌机械工程学报,2024,42(9):865-871.Wang J Q,Hu B,Huang Q,et al.Simulation analysis and experimental verification on bearings fatigue life of a centrifugal pump[J].Journal of Drainage and Irrigation Machinery Engineering,2024,42(9):865-871.
- [12]Rapur J S,Tiwari R.Multifault diagnosis of combined hydraulic and mechanical centrifugal pump faults using continuous wavelet transform and support vector machines[J].Journal of Dynamic Systems,Measurement,and Control,2019,141(11):111013.
- [13]Kamiel B,McKee K,Entwistle R,et al.Multi fault diagnosis of the centrifugal pump using the wavelet transform and principal component analysis[C]//Proceedings of the 9th IFToMM International Conference on Rotor Dynamics.Cham:Springer International Publishing,2015:555-566.
- [14]Rapur J S,Tiwari R.A robust algorithm to detect multiple centrifugal pump faults with corrupted vibration and current signatures using continuous wavelet transform[J].Engineering,2018:203175270.
- [15]张鑫宇,付强,黄倩,等.基于CWT-CNN的离心泵轴承故障识别方法[J].机床与液压,2024,52(12):202-207.Zhang X Y,Fu Q,Huang Q,et al.Fault identification method of centrifugal pump bearing based on CWT-CNN[J].Machine Tool&Hydraulics,2024,52(12):202-207.
- [16]何鹏飞,黄国勇,阮爱国.基于连续小波变换和模型无关元学习的压燃式活塞发动机气门故障诊断研究[J].内燃机工程,2024,45(1):57-63.He P F,Huang G Y,Ruan A G.Study on valve fault diagnosis of compression ignition piston engine based on continuous wavelet transform and model agnostic meta learning[J].Journal of Internal Combustion Engine Engineering,2024,45(1):57-63.
- [17]刘博星,卢金玲,冯建军,等.阀板尾迹对离心泵性能的影响[J].农业工程学报,2024,40(2):217-227.Liu B X,Lu J L,Feng J J,et al.Influence of valve plate wake on the performance of centrifugal pump[J].Transactions of the Chinese Society of Agricultural Engineering,2024,40(2):217-227.
- [18]魏云毅,赵存生,崔哲.离心泵口环间隙对外特性和激励特性的影响[J].中国舰船研究,2021,16(3):189-193.Wei Y Y,Zhao C S,Cui Z.Influence of various wear ring clearances of centrifugal pump on output characteristic and excitation force properties[J].Chinese Journal of Ship Research,2021,16(3):189-193.
- [19]Randall R B.Vibration-based condition monitoring:Industrial,automotive and aerospace applications[M].New Jersey:John Wiley&Sons,2021.
- [20]杨飞翔,吴宏飞,赖晓锋,等.浮动式口环几何尺寸对离心泵性能的影响[J].流体机械,2025,53(7):47-52.Yang F X,Wu H F,Lai X F,et al.Effect of geometric dimensions of floating wear rings on centrifugal pump performance[J].Fluid Machinery,2025,53(7):47-52.
- [21]Cheng W,Zhou Q L,Wu S M,et al.EI-ISOA-VMD:Adaptive denoising and detrending method for nuclear circulating water pump impeller[J].Measurement,2025,242:115890.
- [22]Li S,Liu S,Wang J J,et al.Adaptive wavelet threshold function-based M2M Gaussian noise removal method[J].IEEE Internet of Things Journal,2024,11(20),33177-33192.
- [23]纪俊卿,张亚靓,孟祥川,等.自适应小波阈值滚动轴承故障振动信号降噪方法[J].哈尔滨理工大学学报,2021,26(2):124-130.Ji J Q,Zhang Y L,Meng X C,et al.Adaptive wavelet threshold rolling bearing fault vibration signal denoising method[J].Journal of Harbin University of Science and Technology,2021,26(2):124-130.
- [24]Xu S Y,Gong W Q.Analysis of noise source in a centrifugal pump based on vortex sound theory[J].Physics of Fluids,2025,37(1):015197.
- [25]陈昊,郭文普,康凯,等.基于卷积自适应降噪网络的自动调制识别方法[J].无线电工程,2025,55(2):291-297.Chen H,Guo W P,Kang K,et al.Automatic modulation recognition method based on convolutional adaptive denoising network[J].Radio Engineering,2025,55(2):291-297.
- [26]侯云海,任钰.一种改进的小波阈值去噪算法[J].长春工业大学学报,2023,44(4):345-352.,Hou Y HRen Y.Study on a denoising algorithm of improved wavelet threshold[J].Journal of Changchun University of Technology,2023,44(4):345-352.
- [27]陶付东,智一凡,李怀瑞,等.基于多尺度卷积神经网络和门控循环单元的离心泵叶轮故障诊断[J].机电工程,2025,42(5):885-893.Tao F D,Zhi Y F,Li H R,et al.Fault diagnosis of centrifugal pump impeller based on MCNN and GRU[J].Journal of Mechanical&Electrical Engineering,2025,42(5):885-893.
- [28]斯里达尔·阿拉,苏曼·卡拉扬·阿达里.Python深度学习异常检测使用Keras和PyTorch[M].杨小冬,译.北京:清华大学出版社,2020.Alla S,Adari S K.Beginning anomaly detection using Python-based deep learning with Keras and PyTorch[M].Translated by Yang X D.Beijing:Tsinghua University Press,2020.
- [29]Wisyaldin M K,Luciana G M,Pariaman H.Using LSTM network to predict circulating water pump bearing condition on coal fired power plant[C]//2020 International Conference on Technology and Policy in Energy and Electric Power (ICT-PEP).IEEE,2020:54-59.
- [30]Bukhtoyarov V V,Tynchenko V S,Bashmur K A,et al.Research of methods for pumping technological equipment condition predictive monitoring[C]//Journal of Physics:Conference Series.IOP Publishing,2019,1399(5):055082.
- [31]成功.基于深度学习的换流站阀冷系统主循环泵的故障诊断研究[D].太原:太原理工大学,2022.Cheng G.Research on fault diagnosis of main circulating pump in valve cooling system of converter station based on deep learning[D].Taiyuan:Taiyuan University of Technology,2022.
- [32]Dallapiccola D.Predictive maintenance of centrifugal pumps:A neural network approach[D].Madrid:Escuela Técnica Superior de Ingenieros Informáticos,Universidad Politécnica de Madrid,2020.
- [33]卢俊泽.基于数据驱动的离心泵故障诊断与预测方法研究[D].淮安:淮阴工学院,2024.Lu J Z.Research on centrifugal pump fault diagnosis and prediction methods based on data-driven[D].Huaian:Huaiyin Institute of Technology,2024.
- 循环泵
- 口环磨损
- 自适应小波阈值降噪(AWTD)
- 长短期记忆神经网络(LSTM)
- 趋势预测
circulating pump - wear of mouth ring
- adaptive wavelet threshold denoising(AWTD)
- long short-term memory(LSTM)
- trend prediction