Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2021, Vol. 55 Issue (10): 1856-1866    DOI: 10.3785/j.issn.1008-973X.2021.10.007
计算机技术     
面向工业平稳/非平稳复杂系统的在线故障监测技术
孔祥玉(),王晓兵,李红增,罗家宇
火箭军工程大学 导弹工程学院，陕西 西安 710000
On-line fault monitoring technology for industrial stationary/nonstationary complex system
Xiang-yu KONG(),Xiao-bing WANG,Hong-zeng LI,Jia-yu LUO
Department of Missile Engineering, Rocket Force University of Engineering, Xi’an 710000, China
 全文: PDF(1586 KB)   HTML
摘要：

针对复杂系统中关键性能指标（KPI）相关故障检测方法检测精度低的问题，提出基于双层改进潜结构投影（DL-IPLS）的KPI相关故障检测方法. 利用协整分析和主元分析建立底层模型，对非平稳和平稳变量进行特征提取. 将提取的信息进行融合，建立改进潜结构投影的上层模型，根据融合信息对KPI的贡献进行空间分解. 在2个正交子空间中设计统计量，实现KPI相关故障的在线监测. 田纳西-伊斯曼过程和青霉素发酵过程的仿真结果表明，在面向工业平稳和非平稳复杂工业系统检测时，所提方法有效提高了KPI相关故障的检测率，降低了KPI无关故障的误报率.
关键词： 数据驱动;  协整分析;  改进潜结构投影;  关键性能指标;  过程监控;  故障检测    
Abstract:

A KPI related fault detection method based on double-layer improved projection to latent structures (DL-IPLS) was proposed in view of the low detection accuracy of key performance indicators (KPI) related fault detection methods in complex systems. The underlying model was established by using cointegration analysis and principal component analysis in order to extract the features of non-stationary and stationary variables. Then the extracted information was fused to establish the upper model of improved projection to latent structures. Statistics were designed in two orthogonal subspaces to realize on-line monitoring of KPI related faults. The simulation results of Tennessee-Eastman process and penicillin fermentation process show that the proposed method can effectively improve the detection rate of KPI related faults for industrial stationary and non-stationary complex industrial systems, and reduce the false alarm rate of KPI independent fault.
Key words: data driven    cointegration analysis    improved projection to latent structures    key performance indicator    process monitoring    fault detection
收稿日期: 2021-01-23 出版日期: 2021-10-27
CLC:  TP 277  
基金资助: 国家自然科学基金资助项目（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，从事复杂系统故障诊断研究. orcid.org/0000-0003-2084-7826. E-mail： xiangyukong@126.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
王晓兵
李红增
罗家宇

引用本文:

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.

Xiang-yu KONG,Xiao-bing WANG,Hong-zeng LI,Jia-yu LUO. On-line fault monitoring technology for industrial stationary/nonstationary complex system. Journal of ZheJiang University (Engineering Science), 2021, 55(10): 1856-1866.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2021.10.007    或    https://www.zjujournals.com/eng/CN/Y2021/V55/I10/1856

图 1  基于DL-IPLS故障监测流程图
监测算法 ${J_{{\rm{th,}}{T}_{\hat x}^2}}$ ${J_{{\rm{th,}}{T}_{\tilde x}^2}}$
MPLS 6.70 58.58
OSC-MPLS 7.10 52.75
IPLS 11.53 55.46
DL-IPLS 17.24 20.71
表 1  4种算法在TE过程中的控制限
故障编号 KPI相关故障描述 FDR /%
MPLS OSC-MPLS IPLS DL-IPLS
IDV（1） A/C进料流量比发生变化，成分B恒定 89.87 89.00 99.50 99.65
IDV（2） 组分B质量浓度发生变化，A/C供料比恒定 88.12 96.75 94.37 98.50
IDV（5） 冷凝器冷却水的入口温度 100 100 25.25 24.38
IDV（6） A供料损失（管道1） 99.12 98.75 99.12 99.13
IDV（7） C存在压力损失 41.12 49.37 73.00 99.63
IDV（8） 物料A，B，C供料质量浓度（管道4） 67.12 77.37 94.75 90.88
IDV（10） 物料C供料温度发生变化 46.00 54.37 46.12 77.87
IDV（12） 压缩机冷凝水入口温度变化 84.12 86.37 93.37 91.13
IDV（13） 反应器中的反应程度 90.37 89.87 90.12 92.38
IDV（17） 未知 54.87 65.37 56.50 69.25
IDV（18） 未知 90.00 89.75 88.50 89.13
IDV（20） 未知 49.25 68.25 39.25 75.88
IDV（21） 阀固定在稳态位置 70.25 75.87 56.62 34.50
表 2  TE过程中KPI相关故障有效报警率
图 2  DL-IPLS对KPI相关故障IDV（5）的检测结果
图 3  DL-IPLS对KPI相关故障IDV（7）的检测结果
图 4  OSC-MPLS对KPI相关故障IDV（10）的检测结果
图 5  IPLS对故障KPI相关IDV（10）的检测结果
图 6  DL-IPLS对KPI相关故障IDV（10）的检测结果
故障编号
KPI无关故障描述 FAR /%
MPLS OSC-MPLS IPLS DL-IPLS
IDV（3） 物料D供料温度发生变化 13.62 13.50 7.87 1.37
IDV（4） 反应器冷却水入口温度变化 11.00 9.37 22.00 1.37
IDV（9） D供料温度发生变化 7.50 5.00 9.12 1.12
IDV（11） 反应冷却水入口温度变化 9.62 8.50 21.37 3.37
IDV（15） 压缩机冷凝水阀门 10.51 10.75 13.25 0.62
IDV（16） 未知 45.86 35.00 28.38 66.00
IDV（19） 未知 7.00 10.88 3.38 1.25
表 3  TE过程中KPI无关故障误报警率
图 7  OSC-MPLS对KPI无关故障IDV（4）的检测结果
图 8  IPLS对KPI无关故障IDV（4）的检测结果
图 9  DL-IPLS对KPI无关故障IDV（4）的检测结果
标号 变量 标号 变量
1 采样时间 10 反应器体积
2 通风速率 11 排气二氧化碳浓度
3 搅拌速率 12 PH值
4 底物流加速率 13 温度
5 补料温度 14 产生热
6 底物物质的量 15 酸流加速率
7 溶解氧物质的量 16 碱流加速率
8 菌体物质的量 17 冷水流加速率
9 产物物质的量 18 热水流加速率
表 4  Pensim2.0 仿真数据变量
图 10  DL-IPLS对青霉素发酵过程中故障1的检测结果
图 11  DL-IPLS对青霉素发酵过程中故障2的检测结果
1 LIU J, WANG D, CHEN J Monitoring framework based on generalized tensor PCA for three-dimensional batch process data[J]. Industrial and Engineering Chemistry Research, 2020, 59 (22): 10493- 10508
doi: 10.1021/acs.iecr.9b06244
2 罗林, 苏宏业, 班岚 Dirichlet过程混合模型在非线性过程监控中的应用[J]. 浙江大学学报: 工学版, 2015, 49 (11): 2230- 2236
LUO Lin, SU Hong-ye, BAN Lan Nonparametric bayesian based on mixture of dirichlet process in application of fault detection[J]. Journal of Zhejiang University: Engineering Science, 2015, 49 (11): 2230- 2236
3 LI Z, YAN X Performance-driven ensemble ICA chemical process monitoring based on fault-relevant models[J]. Soft Computing, 2020, 24 (16): 12289- 12302
doi: 10.1007/s00500-020-04673-6
4 LI Z, YAN X Fault-relevant optimal ensemble ICA model for non-gaussian process monitoring[J]. IEEE Transactions on Control Systems Technology, 2020, 28 (6): 2581- 2590
doi: 10.1109/TCST.2019.2936793
5 ZHONG B, WANG J, ZHOU J L, et al Quality rela-ted statistical process monitoring method based on gl-obal and local partial least squares projection[J]. In-dustrial and Engineering Chemistry Research, 2016, 55 (6): 1609- 1622
doi: 10.1021/acs.iecr.5b02559
6 WU O, BOUASWAIG A, IMSLAND L, et al Camp-aign-based modeling for degradation evolution in bat-ch processes using a multiway partial least squares a-pproach[J]. Computers and Chemical Engineering, 2019, 128 (2): 117- 127
7 LIU H, YANG J, ZHANG Y, et al Monitoring of wastewater treatment processes using dynamic concurrent kernel partial least squares[J]. Process Safety and Environmental Protection, 2021, 147: 274- 282
doi: 10.1016/j.psep.2020.09.034
8 LI G, QIN S J, JI Y D Total PLS based contribution plots for fault diagnosis[J]. Acta Automatica Sinica, 2009, 35 (6): 759- 765
9 ZHOU D H, LI G, QIN S J Total projection to latent structures for process monitoring[J]. AiChE Journal, 2010, 56 (1): 168- 178
10 QIN S J, ZHANG Y 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
11 YIN S, DING S X, ZHANG P, et al Study on modifications of PLS approach for process monitoring[J]. IFAC Proceedings Volumes, 2011, 44 (1): 12389- 12394
doi: 10.3182/20110828-6-IT-1002.02876
12 YIN S, ZHU X P, KAYNAK O Improved PLS focused on key performance indictor related fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2015, 62 (3): 1651- 1658
doi: 10.1109/TIE.2014.2345331
13 WANG G, YIN S Quality-related fault detection approach based on orthogonal signal correction and modified PLS[J]. IEEE Transactions on Industrial Informatics, 2017, 11 (2): 398- 405
14 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.
15 CHEN Q, KRUGER U, LEUNG A Y T Cointegration testing method for monitoring nonstationary processes[J]. Industrial and Engineering Chemistry Research, 2009, 48 (7): 3533- 3543
doi: 10.1021/ie801611s
16 SUN H, ZHANG S M, ZHAO C H, et al A sparse reconstruction strategy for online fault diagnosis in nonstationary processes with no a priori fault information[J]. Industrial and Engineering Chemistry Research, 2017, 56 (24): 6993- 7008
doi: 10.1021/acs.iecr.7b00156
17 ZHAO C H, HUANG B A full condition monitoring method for non-stationary dynamic chemical processes with cointegration and slow feature analysis[J]. ALCHE Journal, 2018, 64 (5): 1662- 1681
doi: 10.1002/aic.16048
18 LIN Y, KRUGER U, CHEN Q Monitoring nonstationary dynamic systems using cointegration and common trends analysis[J]. Industrial and Engineering Chemistry Research, 2017, 56 (31): 8895- 8905
doi: 10.1021/acs.iecr.7b00011
19 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
20 JOHANSEN S, JUSELIUS K Maximum likelihood estimation and inference on cointegration with applications to the demand for money[J]. Oxford Bulletin of Economics and Statistics, 1990, 52 (2): 169- 210
21 ZHAO C H, SUN H Dynamic distributed monitoring strategy for large-scale nonstationary processes subject to frequently varying conditions under closed-loop control[J]. IEEE Transactions on Industrial Electronics, 2019, 66 (6): 4749- 4758
doi: 10.1109/TIE.2018.2864703
22 TABRIZI A A, AL-BUGHARBEE H, TRENDAFILOVA I, et al A cointegration-based monitoring method for rolling bearings working in time-varying operational conditions[J]. Meccanica, 2017, 52 (4-5): 1201- 1217
doi: 10.1007/s11012-016-0451-x
23 WILMS I, CROUX C Forecasting using sparse cointegration[J]. International Journal of Forecasting, 2016, 32 (4): 1256- 1267
doi: 10.1016/j.ijforecast.2016.04.005
24 LI G, QIN S J, TAO Y Non-stationarity and cointegration tests for fault detection of dynamic processes[J]. IFAC Proceedings Volumes, 2014, 47 (3): 10616- 10621
doi: 10.3182/20140824-6-ZA-1003.00754
25 BIROL G, ÜNDEY C, CINAR A A modular simulation package for fed-batch fermentation: penicillin production[J]. Computers and Chemical Engineering, 2002, 26 (11): 1553- 1565
doi: 10.1016/S0098-1354(02)00127-8
26 DOWNS J J, VOFEL E F A plant-wide industrial pr-ocess control problem[J]. Computers and Chemical Engineering, 1993, 17 (3): 245- 255
doi: 10.1016/0098-1354(93)80018-I
27 张成, 高宪文, 李元 基于k近邻主元得分差分的故障检测策略[J]. 自动化学报, 2020, 46 (10): 2229- 2238
ZHANG Cheng, GAO Xian-wen, LI Yuan Fault detection strategy based on principal component score difference of k nearest neighbors[J]. Acta Automatica Sinica, 2020, 46 (10): 2229- 2238
[1] 章超波,刘永政,李宏波,赵阳,张丽珠,王子豪. 基于加权残差聚类的建筑负荷预测区间估计[J]. 浙江大学学报(工学版), 2022, 56(5): 930-937.
[2] 宋家成,闫茂德,杨盼盼,巨永锋,岳靖斐. 基于数据驱动的鲁棒反步自适应巡航控制[J]. 浙江大学学报(工学版), 2022, 56(3): 485-493.
[3] 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.
[4] 蔡畅,黄亦翔,邢宏文. 基于梯度提升树的飞机机身对接状态识别[J]. 浙江大学学报(工学版), 2019, 53(7): 1274-1281.
[5] 张燕,王建宙,李威,王婕,陈玲玲,杨鹏. 基于数据驱动的膝关节外骨骼控制[J]. 浙江大学学报(工学版), 2019, 53(10): 2024-2033.
[6] 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.
[7] 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.
[8] 宋筱轩,冯天恒,黄平捷,侯迪波,张光新. 基于动态数据驱动的突发水污染事故仿真方法[J]. 浙江大学学报(工学版), 2015, 49(1): 63-68.
[9] 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.
[10] 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.
[11] 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.
[12] 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.
[13] 潘张鑫 陈偕雄. 或-符合型通用逻辑门组合电路的故障检测[J]. J4, 2007, 41(8): 1260-1264.
[14] 刘世成 王海清 李平. 基于多向核主元分析的青霉素生产过程在线监测[J]. J4, 2007, 41(2): 202-207.
[15] 孟庆华 周晓军 庞茂 潘晓斌 陈明. 疫苗算法及其在车辆故障检测中的应用[J]. J4, 2006, 40(2): 298-303.