解剑波,范海东,李清毅,刘梦杰,赵春晖
XIE Jianbo,FAN Haidong,LI Qingyi,LIU Mengjie,ZHAO Chunhui

• 1:浙江省能源集团有限公司
• 2:浙江大学控制科学与工程学院

摘要(Abstract)：

在自动控制回路的运行过程中,阀门可能会出现粘滞,导致回路发生振荡,控制性能下降。针对阀门粘滞检测问题,提出一种基于全卷积网络的阀门粘滞自动检测方法。使用仿真模型生成阀门正常和出现粘滞时的控制器输出以及过程变量数据,并以此对全卷积网络进行训练。在线应用时,将待测回路的运行数据经过预处理后输入训练好的网络中,以得到阀门粘滞检测结果。在某火电厂的密封水控制回路数据以及公开数据集上的实验结果表明,该方法能够准确地进行阀门粘滞检测。
During the operation of the automatic control loop, the valve may suffer from stiction, which will cause loop oscillation and control performance degradation. Therefore, this paper proposes an automatic valve stiction detection method based on fully convolutional network. A simulation model is used to generate the controller output and process variable data during normal valve operation and stiction, and the data are used to train the fully convolutional network. In the online application stage, the operating data of the loop to be tested is preprocessed and then input into the trained network to get the valve stiction detection result. Experimental results on the sealing water control loop data of a thermal power plant and public data set show that this method can accurately detect valve stiction.

关键词(KeyWords)： 阀门粘滞;全卷积网络;多变量时间序列
valve stiction;fully convolutional network;multivariate time series

Abstract:

Keywords:

基金项目(Foundation): 浙江省重点研发项目（2019C01048）

作者(Author): 解剑波,范海东,李清毅,刘梦杰,赵春晖
XIE Jianbo,FAN Haidong,LI Qingyi,LIU Mengjie,ZHAO Chunhui

DOI: 10.19585/j.zjdl.202102018

参考文献(References)：

• [1]李晨.控制回路的性能监控、诊断与改善[D].上海：华东理工大学，2015.
• [2]DI CAPACI R B,SCALI C.Review and comparison of techniques of analysis of valve stiction:From modeling to smart diagnosis[J].Chemical Engineering Research and Design,2018(130):230-265.
• [3]HORCH A.A simple method for detection of stiction in control valves[J].Control Engineering Practice,1999,7(10):1221-1231.
• [4]郑丽丽，王志国，刘飞.模糊聚类下气动阀门粘滞检测方法的改进[J].机械科学与技术，2018,37(2):300-305.
• [5]CHOUDHURY M A A S,SHAHS L,THORNHILL N F,et al.Automatic detection and quantification of stiction in control valves[J].Control engineering practice,2006,14(12):1395-1412.
• [6]CHOUDHURY M A A S,SHAH S L,THORNHILL N F.Diagnosis of poor control-loop performance using higherorder statistics[J].Automatica,2004,40(10):1719-1728.
• [7]HE Q P,WANG J,POTTMANN M,et al.A curve fitting method for detecting valve stiction in oscillating control Loops[J].Industrial&Engineering Chemistry Research,2007,46(13):4549-4560.
• [8]AMIRUDDIN A A A M,ZABIRI H,TAQVI S A A,et al.Neural network applications in fault diagnosis and detection:an overview of implementations in engineering-relat ed systems[J].Neural Computing and Applications,2020(32):447-472.
• [9]YU W,ZHAO C.Online fault diagnosis for industrial processes with Bayesian network-based probabilistic ensemble learning strategy[J].IEEE Transactions on Automation Science and Engineering,2019,16(4):1922-1932.
• [10]FENG L,ZHAO C.Fault description based attribute transfer for zero-sample industrial fault diagnosis[J].IEEE Trans.Ind.Informat.,early access,Apr.20,2020,doi:10.1109/TII.2020.2988208.
• [11]DAMBROS J W V,FARENZENA M,TRIERWEILER J O.Oscillation detection and diagnosis in process indus tries by pattern recognition technique[J].IFAC-PapersOnLine,2019,52(1):299-304.
• [12]HENRY Y Y S,ALDRICH C,ZABIRI H.Detection and severity identification of control valve stiction in industrial loops using integrated partially retrained CNN-PCA frameworks[J].Chemometrics and Intelligent Laboratory Systems,2020(206):104-143.
• [13]YU W,ZHAO C.Broad convolutional neural network based industrial process fault diagnosis with incremental learning capability[J].IEEE Transactions on Industrial Electronics,2019,67(6):5081-5091.
• [14]CHAI Z,ZHAO C.A Fine-grained adversarial network method for cross-domain industrial fault diagnosis[J].IEEE Transactions on Automation ence and Engineering,2020,17(3):1432-1442.
• [15]YU W,ZHAO C.Robust monitoring and fault isolation of nonlinear industrial processes using denoising autoencoder and elastic net[J].IEEE Transactions on Control Systems Technology,2019,28(3):1083-1091.
• [16]BRASIO A S R,ROMANENKO A,FERNDES N C P.Modeling,detection and quantification,and compensation of stiction in control loops:The state of the art[J].Industrial&Engineering Chemistry Research,2014,53(39):15020-15040.
• [17]CHEN S L,TAN K K,HUANG S.Two-layer binary tree data-driven model for valve stiction[J].Industrial&engineering chemistry research,2008,47(8):2842-2848.
• [18]BAUER M,HORCH A,XIE L,et al.The current state of control loop performance monitoring-a survey of applica tion in industry[J].Journal of Process Control,2016(38):1-10.
• [19]AHMED S,SHAH S L,HUANG B.Maintenance issues in oil and gas processes:Detection of valve stiction[C]//Proceedings of the 1st Annual Gas Processing Symposium.Elsevier,2009.
• [20]CHOUDHURY M A A S,THORNHILL N F,SHAH S L.Modelling valve stiction[J].Control engineering practice,2005,13(5):641-658.
• [21]AMIRUDDIN A A A M,ZABIRI H,JEREMIAH S S,et al.Valve stiction detection through improved pattern recog nition using neural networks[J].Control Engineering Practice,2019(90):63-84.
• [22]CHOUDHURY M A A S,THORNHILL N F,SHAH S L.Modelling valve stiction[J].Control Engineering Practice,2005,13(5):641-658.
• [23]GAO X,YANG F,SHANG C,et al.A review of control loop monitoring and diagnosis:Prospects of controller maintenance in big data era[J].Chinese Journal of Chemical Engineering,2016,24(8):952-962.
• [24]FAWAZ H I,FORESTIER G,WEBER J,et al.Deep learning for time series classification:a review[J].Data Mining and Knowledge Discovery,2019,33(4):917-963.
• [25]WANG Z,YAN W,OATES T.Time series classification from scratch with deep neural networks:A strong baseline[C]//2017 International joint conference on neural networks(IJCNN).IEEE,2017.
• [26]LIN M,CHEN Q,YAN S.Network in network[EB/OL].(2013-11-15)[2020-10-01].http://www.researchgate.net/publication/259312908.html.
• [27]KINGMA D P,BA J.Adam:A method for stochastic optimization[EB/OL].(2014-11-10)[2020-10-01].http:www.researchgate.net/publication/269935079.html.
• [28]JELALI M,HUANG B.Detection and diagnosis of stiction in control loops:state of the art and advanced methods[M].London:Springer London,2010.
• [29]TEH W K,ZABIRI H,SAMYUDIA Y,et al.An improved diagnostic tool for control valve stiction based on nonlinear principle component analysis[J].Industrial&Engineering Chemistry Research,2018,57(33):11350-11365.