摘要
为及时、准确地做出故障诊断,本课题采用独立元分析(ICA)和主成分分析(PCA)两种常用的多元统计分析方法对制浆造纸废水处理过程中的传感器故障进行检测并对诊断效果进行对比。结果表明,对于制浆造纸废水数据中偏移和漂移两种故障,ICA模型的故障检测率分别为24%与54%,PCA模型的故障检测率分别为14%和42%,ICA模型的两种故障检测率均高于PCA模型,但是两种模型均无法达到满意的检测效果;对于完全失效故障,ICA和PCA模型的故障检测率均达到100%。
To monitor and control papermaking wastewater treatment process(WWTP) effectively, two common methods of multivariate statistical analysis named independent component analysis(ICA) and principal component analysis(PCA) were used to detect the sensor faults in a papermaking WWTP. The results showed that the detection rates of the bias and drifting faults using ICA were 24% and 54%, respectively. Meanwhile, the bias and drifting faults detection rates of PCA were 14% and 42%. The fault detection rates of ICA were higher than those of PCA, but neither of the two methods achieved satisfactory result of detecting the bias and drifting faults. Concerning the complete failure fault, both the fault detection rates of the two methods were 100%.
引文
[1] Ge Z,Song Z,Gao F.Review of recent research on data-based process monitoring[J].Industrial & Engineering Chemistry Research,2013,52(10):3543.
[2] Liu T L,Shen W H.A review of application of fault diagnostic expert system in wastewater treatment[J].Paper Science & Technology,2011,30(2):75.刘天龙,沈文浩.污水处理过程中故障诊断专家系统的应用[J].造纸科学与技术,2011,30(2):75.
[3] Huang D P,Qiu Y,Liu Y Q,et al.Review of data-driven fault diagnosis and prognosis for wastewater treatment[J].Journal of South China University of Technology (Natural Science Edition),2015,43(3):111.黄道平,邱禹,刘乙奇,等.面向污水处理的数据驱动故障诊断及预测方法综述[J].华南理工大学学报(自然科学版),2015,43(3):111.
[4] Zhou D H,Hu Y Y.Fault diagnosis techniques for dynamic systems[J].Acta Automatic Sinica,2009,35(6):748.周东华,胡艳艳.动态系统的故障诊断技术[J].自动化学报,2009,35(6):748.
[5] Lee J M,Yoo C K,Lee I B.Statistical process monitoring with independent component analysis[J].Journal of Process Control,2004,14(5):467.
[6] Xie Z H.Matlab statistical analysis and application of 40 case studies[M].Beijing:Beihang University Press,2010.谢中华.MATLAB统计分析与应用:40个案例分析[M].北京:北京航空航天大学出版社,2010.
[7] Qin S J.Survey on data-driven industrial process monitoring and diagnosis[J].Annual Reviews in Control,2012,36(2):220.
[8] Shen Y,Ding S X,Haghani A,et al.A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process[J].Journal of Process Control,2012,22(9):1567.
[9] Qin S J.Statistical process monitoring:basics and beyond[J].Journal of Chemometrics,2003,17(8/9):480.
[10] Olsson G.ICA and me—A subjective review[J].Water Research,2012,46(6):1585.
[11] Hyv?rinen A,Oja E.Independent component analysis:algorithms and applications[J].Neural Networks,2000,13(5):411.
[12] Hyv?rinen A.Fast and robust fixed-point algorithms for independent component analysis[J].IEEE Transactions on Neural Networks,1999,10(3):626.
[13] Cardoso F,Souloumiac A.Blind beamforming for non Gaussian signals[J].Radar & Signal Processing IEE Proceedings F,1993,140(6):362.
[14] Silverman B W.Density estimation for statistics and data analysis[M].London:Chapman & Hall,1986.
[15] Liu Y Y,Li X Y,Zhang G R,et al.Sensor fault detection of papermaking wastewater treatment processses based on multivariate statistical analysis[J].China Pulp & Paper Industry,2017,38(8):41.刘耀瑶,李祥宇,张光锐,等.基于多元统计分析的造纸废水处理过程传感器故障检测[J].中华纸业,2017,38(8):41.
[16] Dunia R,Qin S J,Edgar T F,et al.Identification of faulty sensors using principal component analysis[J].AICHE Journal,1996,42(10):2797.