Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2023, Vol. 57 Issue (7): 1297-1306    DOI: 10.3785/j.issn.1008-973X.2023.07.004
自动化技术     
基于动态内并行潜结构投影的故障检测方法
孔祥玉1(),陈雅琳1,2,罗家宇1,杨治艳3
1. 火箭军工程大学 导弹工程学院，陕西 西安 710025
2. 航空工业成都凯天电子股份有限公司，四川 成都 610091
3. 工业和信息化部电子第五研究所，广东 广州 511370
Fault detection method based on dynamic inner concurrent projection to latent structure
Xiang-yu KONG1(),Ya-lin CHEN1,2,Jia-yu LUO1,Zhi-yan YANG3
1. College of Missile Engineering, Rocket Force University of Engineering, Xi’an 710025, China
2. AVIC Chengdu Caic Electronics Limited Company, Chengdu 610091, China
3. China Electronic Product Reliability and Environmental Testing Research Institute, Guangzhou 511370, China
 全文: PDF(1047 KB)   HTML
摘要：

针对实际工业过程故障检测时存在误报警现象及易缺失部分时段质量数据的问题，提出在线监测动态内潜结构投影(OM-DiPLS)模型. 该模型通过引入时延的质量数据，使得在缺失部分时段质量数据时能够实现模型的更新. 为了更好地监控质量变量中不可预测的信息，基于OM-DiPLS模型提出在线监测动态内并行潜结构投影模型. 该模型将过程数据和质量数据投影到输入输出相关的协变子空间、输出无关但过程相关的输入主子空间、输入残差子空间、不可预测的输出主子空间及输出残差子空间，通过对各子空间构造相应的统计量，实现过程监测. 田纳西-伊斯曼过程仿真的实验表明，利用所提算法有效提高了质量相关故障的有效检测率，降低了质量无关故障的误报率.
关键词： 偏最小二乘;  关键性能指标;  动态建模;  过程监测;  故障检测    
Abstract:

The online monitoring dynamic inner projection to latent structures (OM-DiPLS) model was proposed aiming at the problems of false alarm and missing quality data of some periods in actual industrial process fault detection. The model can be updated when quality data of some periods were missing by introducing quality data of time delay. An online monitoring dynamic internal concurrent latent structure projection (OM-DiCPLS) model was proposed based on the OM-DiPLS model in order to better monitor the unpredictable information in the quality variables. The process data and quality data were projected into the covariant subspace with relevant inputs and outputs, the input principal subspace with unrelated outputs but relevant processes, the input residual subspace, the unpredictable output principal subspace and the output residual subspace. Process monitoring was realized by constructing corresponding statistics for each subspace. The Tennessee-Eastman process simulation experiment shows that the proposed algorithm effectively improves the effective detection rate of quality-related faults and reduces the false alarm rate of quality-unrelated faults.
Key words: partial least square    key performance indicator    dynamic modeling    process monitoring    fault detection
收稿日期: 2022-05-21 出版日期: 2023-07-17
CLC:  TP 277  
基金资助: 国家自然科学基金资助项目（62273354，61673387，61833016）；陕西省自然科学基金资助项目（2020JM-356）
作者简介: 孔祥玉（1967—），男，教授，博导，从事复杂系统故障诊断的研究. orcid.org/0000-0003-2084-7826. E-mail： xiangyukong01@163.com
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
孔祥玉
陈雅琳
罗家宇
杨治艳

引用本文:

孔祥玉,陈雅琳,罗家宇,杨治艳. 基于动态内并行潜结构投影的故障检测方法[J]. 浙江大学学报(工学版), 2023, 57(7): 1297-1306.

Xiang-yu KONG,Ya-lin CHEN,Jia-yu LUO,Zhi-yan YANG. Fault detection method based on dynamic inner concurrent projection to latent structure. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1297-1306.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2023.07.004    或    https://www.zjujournals.com/eng/CN/Y2023/V57/I7/1297

子空间 统计量 计算 控制限
${S_{{\rm{CVS}}} }$ $T_{\rm{c}}^2$ ${\boldsymbol{v} }_k{\boldsymbol{\varLambda} } _k^{ - 1}{ {\boldsymbol{v} }_k^{\rm{T} } }$ ${J_{ {\rm{th} },T_{\rm{c}}^2} }{\text{ = } }\dfrac{ {A({n^2} - 1)} }{ {n(n - A)} }{F_{A,n - A,\alpha } }$
${S_{{\rm{IPS}}} }$ $ T_x^2 $ ${\boldsymbol{v} }_x{\boldsymbol{\varLambda} } _x^{ - 1}{ {\boldsymbol{v} }_x^{\rm{T} } }$ ${J_{{\rm{th}},T_x^2} } = \dfrac{ { {A_x}({n^2} - 1)} }{ {n(n - {A_x})} }{F_{ {A_x},n - {A_x},\alpha } }$
${S_{{\rm{OPS}}} }$ $ T_y^2 $ ${\boldsymbol{v} }_y{\boldsymbol{\varLambda} } _y^{ - 1}{ {\boldsymbol{v} }_y^{\rm{T}}}$ ${J_{{\rm{th}},T_y^2} }{\text{ = } }\dfrac{ { {A_y}({n^2} - 1)} }{ {n(n - {A_y})} }{F_{ {A_y},n - {A_y},\alpha } }$
${S_{{\rm{IRS}}} }$ $ {Q_x} $ ${\left\| { { {{\boldsymbol{\tilde x}}}_x} } \right\|^2}$ ${J_{{\rm{th}},{Q_x} } } = g\chi _{h,\alpha }^2$
${S_{{\rm{ORS}}} }$ $ {Q_y} $ ${\left\| { { {{\boldsymbol{\tilde y}}}_y} } \right\|^2}$ ${J_{{\rm{th}},{Q_y} } } = g\chi _{h,\alpha }^2$
表 1  OM-DiCPLS模型的统计指标
故障
编号 		DTPLS OMDC-PLS OM-DiCPLS
${S_{\rm{D}}}$ ${S_{{\rm{WD}}} }$ ${S_{{\rm{WG}}} }$ ${S_{{\rm{DT}}} }$ ${S_{{\rm{CVS}}} }$ ${S_{{\rm{IPS}}} }$ ${S_{{\rm{IRS}}} }$ ${S_{{\rm{OPS}}} }$ ${S_{{\rm{ORS}}} }$ ${S_{{\rm{CVS}}} }$ ${S_{{\rm{IPS}}} }$ ${S_{{\rm{IRS}}} }$ ${S_{{\rm{OPS}}} }$ ${S_{{\rm{ORS}}} }$
1 99.75 99.50 99.88 99.88 99.38 99.63 100.0 98.50 06.13 100.0 99.88 100.0 100.0 99.62
2 99.50 98.75 98.50 99.13 98.63 98.88 100.0 97.88 02.88 99.25 99.37 100.0 98.99 99.37
5 36.89 24.06 16.00 46.88 33.13 29.25 4.13 26.75 02.13 28.55 19.87 100.0 58.24 52.45
6 100.0 98.62 99.50 100.0 99.38 99.38 100.0 97.88 53.13 99.37 99.62 100.0 100.0 99.87
7 58.97 97.24 100.0 99.50 61.00 100.0 100.0 39.13 04.13 47.92 100.0 100.0 80.25 67.92
8 93.22 96.87 91.38 93.63 98.75 97.50 100.0 89.25 04.63 97.99 97.48 100.0 98.36 98.87
10 32.12 18.42 11.75 66.13 63.00 29.25 44.78 32.13 01.88 48.18 11.95 99.75 76.86 72.08
12 86.83 96.12 89.00 92.88 97.63 98.63 97.83 82.38 16.75 95.85 97.61 100.0 99.87 98.74
13 92.35 93.86 94.88 95.63 94.63 95.38 100.0 90.25 14.25 93.58 94.88 99.87 97.48 97.36
表 2  质量相关故障检测率
图 1  质量相关故障(12)的监测效果
图 2  OM-DiCPLS方法对故障(2)的监测效果
故障
编号 		DTPLS OMDC-PLS OM-DiCPLS
${S_{\rm{D}}}$ ${S_{{\rm{WD}}} }$ ${S_{{\rm{WG}}} }$ ${S_{{\rm{DT}}} }$ ${S_{{\rm{CVS}}} }$ ${S_{{\rm{IPS}}} }$ ${S_{{\rm{IRS}}} }$ ${S_{{\rm{OPS}}} }$ ${S_{{\rm{ORS}}} }$ ${S_{{\rm{CVS}}} }$ ${S_{{\rm{IPS}}} }$ ${S_{{\rm{IRS}}} }$ ${S_{{\rm{OPS}}} }$ ${S_{{\rm{ORS}}} }$
3 13.84 11.65 6.13 5.75 14.88 5.00 01.74 10.20 00.50 07.42 03.89 98.87 50.69 45.66
4 24.22 14.04 95.63 99.88 14.75 52.00 100.0 09.00 01.25 07.04 25.91 100.0 58.87 50.44
9 14.09 10.78 07.88 7.63 15.88 6.38 02.17 06.38 00.50 08.68 03.77 99.50 49.56 44.40
11 29.36 31.33 69.13 71.13 27.50 64.13 89.35 07.38 00.75 11.70 47.92 100.0 62.52 65.53
14 99.62 100.0 100.0 91.13 4.75 99.88 100.0 06.50 01.00 03.27 100.0 100.0 41.51 38.99
15 14.34 10.03 05.88 8.25 11.25 9.88 01.74 11.00 00.75 08.30 03.90 97.99 46.79 40.63
表 3  质量无关故障的误报率
图 3  质量无关故障(4)的监测效果
1 刘强, 卓洁, 郎自强, 等 数据驱动的工业过程运行监控与自优化研究展望[J]. 自动化学报, 2018, 44 (11): 1944- 1956
LIU Qiang, ZHUO Jie, LANG Zi-qiang, et al Perspectives on data-driven operation monitoring and self-optimization of industrial processes[J]. Acta Automatica Sinica, 2018, 44 (11): 1944- 1956
doi: 10.16383/j.aas.2018.c180207
2 孔祥玉, 王晓兵, 李红增, 等 面向工业平稳/非平稳复杂系统的在线故障监测技术[J]. 浙江大学学报: 工学版, 2021, 55 (10): 1856- 1866
KONG Xiang-yu, WANG Xiao-bing, LI Hong-zeng, et al On-line fault monitoring technology for industrial stationary/nonstationary complex system[J]. Journal of Zhejiang University: Engineering Science, 2021, 55 (10): 1856- 1866
3 孙鹤. 数据驱动的复杂非平稳工业过程建模与监测[D]. 杭州: 浙江大学, 2018: 8−19.
SUN He. Complex nonstationary industrial process modeling and monitoring based on data driven methods [D]. Hangzhou: Zhejiang University, 2018: 8−19.
4 张成, 郭青秀, 李元, 等 基于主元分析得分重构差分的故障检测策略[J]. 控制理论与应用, 2019, 36 (5): 774- 782
ZHANG Cheng, GUO Qing-xiu, LI Yuan, et al Fault detection strategy based on difference of score reconstruction associated with principal component analysis[J]. Control Theory and Applications, 2019, 36 (5): 774- 782
doi: 10.7641/CTA.2018.70915
5 ZENG Y Q, LOU Z J. The new PCA for dynamic and non-Gaussian processes [EB/OL]. [2022-05-05]. https://sci-hub.ee/.
6 顾炳斌, 熊伟丽 基于多块信息提取的PCA故障诊断方法[J]. 化工学报, 2019, 70 (2): 736- 749
GU Bing-bin, XIONG Wei-li Fault diagnosis based on PCA method with multi-block information extraction[J]. CIESC Journal, 2019, 70 (2): 736- 749
doi: 10.11949/j.issn.0438-1157.20180842
7 孔祥玉, 杨治艳, 罗家宇, 等 基于新息矩阵的独立成分分析故障检测方法[J]. 中南大学学报: 自然科学版, 2021, 52 (4): 1232- 1241
KONG Xiang-yu, YANG Zhi-yan, LUO Jia-yu, et al Fault detection method with independent component analysis based on innovation matrix[J]. Journal of Central South University: Science and Technology, 2021, 52 (4): 1232- 1241
8 刘舒锐, 彭慧, 李帅, 等 基于IJB-PCA-ICA算法的故障检测[J]. 化工学报, 2018, 69 (12): 5146- 5154
LIU Shu-rui, PENG Hui, LI Shuai, et al Fault detection based on IJB-PCA-ICA[J]. CIESC Journal, 2018, 69 (12): 5146- 5154
9 孔祥玉, 王晓兵, 罗家宇, 等 基于动态高效潜结构投影的质量相关故障检测[J]. 控制理论与应用, 2021, 38 (12): 2076- 2084
KONG Xiang-yu, WANG Xiao-bing, LUO Jia-yu, et al Quality-related fault detection based on dynamic efficient projection to latent structures[J]. Control Theory and Applications, 2021, 38 (12): 2076- 2084
doi: 10.7641/CTA.2021.00586
10 金鑫, 梁军 基于动态PLS框架的多变量无静差预测控制[J]. 浙江大学学报: 工学版, 2016, 50 (4): 750- 758
JIN Xin, LIANG Jun Multivariable offset free model predictive control in dynamic PLS framework[J]. Journal of Zhejiang University: Engineering Science, 2016, 50 (4): 750- 758
11 GAO X R, SHARDT Y Dynamic system modelling and process monitoring based on long-term dependency slow feature analysis[J]. Journal of Process Control, 2021, 105: 27- 47
doi: 10.1016/j.jprocont.2021.07.007
12 LI G, QIN S J, ZHOU D H Geometric properties of partial least squares for process monitoring[J]. Automatica, 2010, 46 (1): 204- 210
13 ZHOU D H, LI G, QIN S J Total projection to latent structures for process monitoring[J]. AIChE Journal, 2010, 56 (1): 168- 178
14 YIN S, DING S X, ZHANG P, et al Study on modifications of PLS approach for process monitoring[J]. IFAC Proc Volumes, 2011, 44 (1): 12389- 12394
doi: 10.3182/20110828-6-IT-1002.02876
15 PENG K, ZHANG K, YOU B, et al Quality-relevant fault monitoring based on efficient projection to latent structures with application to hot strip mill process[J]. IET Control Theory and Applications, 2015, 9 (7): 1135- 1145
doi: 10.1049/iet-cta.2014.0732
16 QIN S J, ZHENG Y Quality-relevant and process-relevant fault monitoring with concurrent projection to latent structures[J]. AIChE Journal, 2013, 59 (2): 496- 504
doi: 10.1002/aic.13959
17 孔祥玉, 曹泽豪, 安秋生, 等 偏最小二乘线性模型及其非线性动态扩展模型综述[J]. 控制与决策, 2018, 33 (9): 1537- 1548
KONG Xiang-yu, CAO Ze-hao, AN Qiu-sheng, et al Review of partial least squares linear models and their nonlinear dynamic expansion models[J]. Control and Decision, 2018, 33 (9): 1537- 1548
doi: 10.13195/j.kzyjc.2017.1306
18 RICKER N L The use of biased least-squares estimators for parameters in discrete-time pulse-response models[J]. Industrial and Engineering Chemistry Research, 1988, 27 (2): 343- 350
doi: 10.1021/ie00074a023
19 KU W, STORER R H, GEORGAKIS C Disturbance detection and isolation by dynamic principal component analysis[J]. Chemometrics and Intelligent Laboratory Systems, 1995, 30 (1): 179- 196
doi: 10.1016/0169-7439(95)00076-3
20 KASPAR M H, RAY W H Dynamic PLS modelling for process control[J]. Chemical Engineering Science, 1993, 48 (20): 3447- 3461
doi: 10.1016/0009-2509(93)85001-6
21 LAKSHMINARAYANAN S, SHAH S L, NANDAK- UMAR K Modeling and control of multivariable processes: dynamic PLS approach[J]. AIChE Journal, 1997, 43 (9): 2307- 2322
doi: 10.1002/aic.690430916
22 LI G, LIU B S, QIN S J, et al Quality relevant data-driven modeling and monitoring of multivariate dynamic processes: the dynamic T-PLS approach[J]. IEEE Transactions on Neural Networks, 2011, 22 (12): 2262- 2271
doi: 10.1109/TNN.2011.2165853
23 DONG Y, QIN S J Dynamic-inner partial least squares for dynamic data modeling[J]. IFAC-PapersOnLine, 2015, 48 (8): 117- 122
doi: 10.1016/j.ifacol.2015.08.167
24 DONG Y, QIN S J A novel dynamic PCA algorithm for dynamic data modeling and process monitoring[J]. Journal of Process Control, 2018, 67: 1- 11
doi: 10.1016/j.jprocont.2017.05.002
25 DONG Y, QIN S J Dynamic latent variable analytics for process operations and control[J]. Computers and Chemical Engineering, 2018, 114 (9): 69- 80
26 DONG Y, QIN S J Regression on dynamic PLS structures for supervised learning of dynamic data[J]. Journal of Process Control, 2018, 68: 64- 72
doi: 10.1016/j.jprocont.2018.04.006
27 KONG X Y, CAO Z H, AN Q S, et al Quality-related and process-related fault monitoring with online monitoring dynamic concurrent PLS[J]. IEEE Access, 2018, 6: 59074- 59086
doi: 10.1109/ACCESS.2018.2872790
28 QIN S J, DONG Y N, ZHU Q Q, et al Bridging systems theory and data science: a unifying review of dynamic latent variable analytics and process monitoring[J]. Annual Reviews in Control, 2020, 50 (1): 29- 48
29 DOWNS J J, VOGEL E F A plant-wide industrial process control problem[J]. Pergamon, 1993, 17 (3): 245- 255
30 JIANG Q C, YAN X F, LI J, et al PCA-ICA integrated with Bayesian method for non-Gaussian fault diagnosis[J]. Industrial and Engineering Chemistry Research, 2016, 55 (17): 4979- 4986
doi: 10.1021/acs.iecr.5b04023
31 宋凯, 王海清, 李平 PLS质量监控及其在Tennessee Eastman过程中的应用[J]. 浙江大学学报: 工学版, 2005, 39 (5): 657- 662
SONG Kai, WANG Hai-qing, LI Ping PLS quality monitoring and its application for Tennessee Eastman process[J]. Journal of Zhejiang University: Engineering Science, 2005, 39 (5): 657- 662
32 张展博, 王振雷, 王昕 基于正交局部慢性特征的故障检测方法[J]. 清华大学学报: 自然科学版, 2020, 60 (8): 693- 700
ZHANG Zhan-bo, WANG Zhen-lei, WANG Xin A fault detection method based on orthogonal local chronic features[J]. Journal of Tsinghua University: Natural Science Edition, 2020, 60 (8): 693- 700
33 张成, 潘立志, 李元 基于加权统计特征KICA的故障检测与诊断方法[J]. 化工学报, 2022, 73 (2): 827- 837
ZHANG Cheng, PAN Li-zhi, LI Yuan Fault detection and diagnosis method based on weighted statistical feature KICA[J]. CIESC Journal, 2022, 73 (2): 827- 837
34 韩宇, 李俊芳, 高强, 等 基于故障判别增强KECA算法的故障检测[J]. 化工学报, 2020, 71 (3): 1254- 1263
HAN Yu, LI Jun-fang, GAO Qiang, et al Fault detection based on fault discrimination enhanced kernel entropy component analysis algorithm[J]. CIESC Journal, 2020, 71 (3): 1254- 1263
[1] 刘兴,余建波. 注意力卷积GRU自编码器及其在工业过程监控的应用[J]. 浙江大学学报(工学版), 2021, 55(9): 1643-1651.
[2] 刘晓伟,陈赟,张思,陈康. FDM型增材制造中送丝机构动态监测与识别[J]. 浙江大学学报(工学版), 2021, 55(3): 548-554.
[3] 孔祥玉,王晓兵,李红增,罗家宇. 面向工业平稳/非平稳复杂系统的在线故障监测技术[J]. 浙江大学学报(工学版), 2021, 55(10): 1856-1866.
[4] 金鑫, 梁军. 基于动态PLS框架的多变量无静差预测控制[J]. 浙江大学学报(工学版), 2016, 50(4): 750-758.
[5] 贺刘, 姚铮, 崔晓伟, 陆明泉, 郭婧. 基于信号体制的热噪声模型及对RAIM的影响[J]. 浙江大学学报(工学版), 2015, 49(4): 711-716.
[6] 张鑫, 崔晓伟, 冯振明. 基于稀疏故障假设的鲁棒定位方法[J]. 浙江大学学报(工学版), 2015, 49(10): 1924-1928.
[7] 贺刘, 姚铮, 崔晓伟, 陆明泉, 郭婧. 基于信号体制的热噪声模型及对RAIM的影响[J]. 浙江大学学报(工学版), 2014, 48(11): 4-5.
[8] 朱凡,李悦,蒋 凯,叶树明,郑筱祥. 基于偏最小二乘的大鼠初级运动皮层解码[J]. J4, 2013, 47(5): 901-905.
[9] 赵立杰, 柴天佑, 袁德成, 刁晓坤. 增强操作工况识别可靠性的概率PLS-ELM方法[J]. J4, 2013, 47(10): 1747-1752.
[10] 董学锋,戴连奎,黄承伟. 结合PLS-DA与SVM的近红外光谱软测量方法[J]. J4, 2012, 46(5): 824-829.
[11] 赵振,张树有. 基于分步回归的低压断路器分断电流求解技术[J]. J4, 2012, 46(11): 1943-1952.
[12] 曹玉苹, 田学民. 基于信息散度的过程故障检测与诊断[J]. J4, 2010, 44(7): 1315-1320.
[13] 陈美娟, 陈杰勋, 王靖岱, 蒋斌波, 阳永荣. 聚乙烯密度的拉曼光谱检测[J]. J4, 2010, 44(6): 1164-1168.
[14] 祝志博, 王培良, 宋执环. 基于PCASVDD的故障检测和自学习辨识[J]. J4, 2010, 44(4): 652-658.
[15] 林勇, 周晓军, 杨先勇, 等. 基于双谱识别和人工免疫网络的智能故障检测[J]. J4, 2009, 43(10): 1777-1782.