Hierarchical Monitoring for Multi-unit Chemical Processes Based on Local-global Correlation Features
-
摘要: 针对一类多单元化工过程的监测问题, 提出基于局部−整体相关特征的分层故障检测与故障定位方法, 通过表征单元内部变量相关性、单元与单元间相关性、局部单元与过程整体相关性, 对过程运行状态进行判断, 以提升过程监测的准确性与可靠性. 首先, 采用典型相关分析, 通过引入邻域单元相关变量提取每个单元的独有特征和外部相关特征; 其次, 对每个单元的独有特征和所有单元的外部相关特征建立统计模型实现分层故障检测; 然后, 建立单元−变量分层贡献图, 对故障单元以及故障变量实现分层定位. 通过在Tennessee Eastman仿真过程和一个实验室级甘油精馏过程中的应用说明所提分层监测方法的有效性.Abstract: A hierarchical process monitoring method based on local-global correlation features is proposed for a class of multi-unit chemical processes. The process operation status is identified by characterizing the correlation within a local unit, between units, and between the local unit and the whole process, through which the monitoring reliability is enhanced. First, based on canonical correlation analysis, individual characteristics and external correlation characteristics of each unit are extracted by introducing correlated variables from neighboring units; Second, multivariate statistical monitoring models are established for individual characteristics of each unit and external correlation characteristics of all units; Then, the unit-variable hierarchical contribution plot is established to locate the fault units and fault variables hierarchically. The effectiveness of the proposed hierarchical monitoring is demonstrated through applications to the Tennessee Eastman process and a laboratory distillation process.
-
图 1 基于局部−整体相关特征的分层监测设计框架
Fig. 1 Framework of the local-global correlation feature-based hierarchical monitoring
图 3 TE过程故障4的监测效果 ((a)经典CCA监测结果; (b)分层监测整体监测结果; (c)分层监测局部监测效果)
Fig. 3 Monitoring results for the TE fault 4 ((a) Conventional CCA method; (b) Global monitoring using hierarchical method; (c) Local monitoring using hierarchical method)
图 4 TE过程故障4的分层贡献图((a)
$T_{b,{\rm out}}^2$ ; (b)$T_{b,{\rm in}}^2$ ; (c)${Q_b}$ )Fig. 4 Contribution plots for the TE fault 4 ((a)
$T_{b,{\rm out}}^2$ ; (b)$T_{b,{\rm in}}^2$ ; (c)${Q_b}$ )
图 5 TE过程故障5的监测效果 ((a)经典CCA监测结果; (b)分层监测整体监测结果; (c)分层监测局部监测效果)
Fig. 5 Fault detection results for the TE fault 5 ((a) Conventional CCA method; (b) Global monitoring using hierarchical method; (c) Local monitoring using hierarchical method)
图 6 TE过程故障5的分层监测贡献图 ((a)
$T_z^2$ 和${Q_z}$ ; (b)$T_{b,{\rm out}}^2$ ; (c)$T_{b,{\rm in}}^2$ ; (d)${Q_b}$ ; (e)控制补偿后$T_{b,{\rm out}}^2$ )Fig. 6 Contribution plots for the TE fault 5 ((a)
$T_z^2$ and${Q_z}$ ; (b)$T_{b,{\rm out}}^2$ ; (c)$T_{b,{\rm in}}^2$ ; (d)${Q_b}$ ; (e)$T_{b,{\rm out}}^2$ after compensation)
图 7 实验室甘油精馏装置与流程图((a)设备图; (b)流程图)
Fig. 7 Lab-scale glycerin distillation process ((a) Equipment diagram; (b) Simplified flowchart)
图 8 精馏过程故障1的分层故障检测效果 ((a)整体监测统计量; (b)局部监测统计量)
Fig. 8 Fault detection results for the distillation process fault 1 ((a) Global monitoring statistics; (b) Local monitoring statistics)
图 9 精馏过程故障1分层监测贡献图((a)
$T_z^2$ 和${Q_z}$ ; (b)$T_{b,{\rm out}}^2$ ; (c)$T_{b,{\rm in}}^2$ )Fig. 9 Contribution plots for the distillation fault 1 ((a)
$T_z^2$ and${Q_z}$ ; (b)$T_{b,{\rm out}}^2$ ; (c)$T_{b,{\rm in}}^2$ )
图 10 精馏过程故障2的分层故障检测效果((a)整体监测统计量; (b)局部监测统计量)
Fig. 10 Fault detection results for the distillation process fault 2 ((a) Global monitoring statistics; (b) Local monitoring statistics)
表 1 TE过程的典型操作单元和对应变量
Table 1 Operation units and corresponding variables in the TE process
单元 变量描述 变量名称 符号 进料 A 进料 (流1) XMEAS(1) $\boxed1$ D 进料 (流2) XMEAS(2) $\boxed2$ E 进料 (流3) XMEAS(3) $\boxed3$ A 和 C 进料 XMEAS(4) $\boxed4$ D 进料 XMV(1) A 进料流量 XMV(3) E 进料流量 XMV(2) A 和 C 进料流量 XMV(4) 反应器 反应器进料量 XMEAS(6) $\boxed6$ 反应器压力 XMEAS(7) $\boxed7$ 反应器液位 XMEAS(8) $\boxed8$ 反应器温度 XMEAS(9) $\boxed9$ 反应器水温 XMEAS(21) $\boxed{21}$ 反应器冷却水流量 XMV(10) 冷凝器冷却水流量 XMV(11) 分离器 分离器温度 XMEAS(11) $\boxed{11}$ 分离器液位 XMEAS(12) $\boxed{12}$ 分离器压力 XMEAS(13) $\boxed{13}$ 分离器底物流量 XMEAS(14) $\boxed{14}$ 分离器水温度 XMEAS(22) $\boxed{22}$ 分离器液流量 XMV(7) 汽提塔 汽提塔液位 XMEAS(15) $\boxed{15}$ 汽提塔压力 XMEAS(16) $\boxed{16}$ 汽提塔底物流量 XMEAS(17) $\boxed{17}$ 汽提塔温度 XMEAS(18) $\boxed{18}$ 汽提塔蒸汽流量 XMEAS(19) $\boxed{19}$ 汽提塔产物流量 XMV(8) 汽提塔蒸汽阀开度 XMV(9) 压缩 再循环流量 XMEAS(5) $\boxed5$ 排放速度 XMEAS(10) $\boxed{10}$ 压缩机功率 XMEAS(20) $\boxed{20}$ 压缩机再循环阀 XMV(5) 排放阀 XMV(6) 
下载: 导出CSV
表 2 分层监测对于21个故障测试集的监测效果
Table 2 Hierarchical monitoring results for the 21 faults in TE process
编码 单元及过程 进料单元 反应器单元 分离器单元 汽提塔单元 压缩单元 过程整体 故障描述/统计量 $T_{1,{\rm out}}^2$ $T_{1,{\rm in}}^2$ ${Q_1}$ $T_{2,{\rm out}}^2$ $T_{2,{\rm in}}^2$ ${Q_2}$ $T_{3,{\rm out}}^2$ $T_{3,{\rm in}}^2$ ${Q_3}$ $T_{4,{\rm out}}^2$ $T_{4,{\rm in}}^2$ ${Q_4}$ $T_{5,{\rm out}}^2$ $T_{5,{\rm in}}^2$ ${Q_5}$ $T_z^2$ ${Q_z}$ 1 A/C 进料比率, B 成分不变 (阶跃) 0.99 0.31 0.04 0.77 0.26 0.06 0.44 0.04 0.07 1 0.06 0.98 0.17 0.02 0.23 1 1 2 B 成分, A/C 进料比率不变 (阶跃) 0.92 0.02 0.27 0.95 0.22 0.03 0.92 0.14 0.06 0.99 0.06 0.89 0.99 0.01 0.42 0.98 0.98 3 D 的进料温度 (阶跃) 0.01 0.01 0.01 0.32 0.02 0.00 0.14 0.01 0.00 0.20 0 0.01 0 0.01 0.09 0.01 0.02 4 反应器冷却水入口温度 (阶跃) 0.02 0.01 0.02 0.25 0.75 1 0.12 0.00 0.01 0.21 0 0.01 0 0.00 0.01 0.03 0.06 5 冷凝器冷却水入口温度 (阶跃) 0.16 0.03 0.04 0.99 0.09 0.03 0.23 0.01 0.02 1 0.00 0.19 0.07 0.00 0.13 0.22 0.18 6 A 进料损失 (阶跃) 0.99 0.91 1 1 0.98 0.96 0.98 0.82 0.98 0.99 0.96 0.97 0.99 0.92 0.99 0.99 0.99 7 C 存在压力损失 (阶跃) 0.98 1 0.87 0.98 0.22 0.09 0.38 0.03 0.04 0.76 0.01 0.22 0.24 0.01 0.27 1 0.98 8 A、B、C 进料成分 (随机) 0.78 0.10 0.16 0.97 0.48 0.13 0.90 0.03 0.35 0.94 0.11 0.68 0.87 0.02 0.61 0.97 0.89 9 D 的进料温度 (随机) 0.00 0.01 0.01 0.27 0.02 0.01 0.14 0.01 0.01 0.16 0 0.01 0 0.00 0.02 0.01 0.02 10 C 的进料温度 (随机) 0.08 0.02 0.02 0.43 0.04 0.02 0.33 0.01 0.00 0.46 0.00 0.81 0.07 0.00 0.10 0.29 0.13 11 反应器冷却水入口温度 (随机) 0.11 0.01 0.01 0.39 0.61 0.70 0.17 0.01 0.01 0.42 0.00 0.05 0 0.01 0.02 0.20 0.27 12 冷凝器冷却水入口温度 (随机) 0.74 0.22 0.25 0.95 0.60 0.29 0.94 0.28 0.65 0.96 0.06 0.89 0.34 0.03 0.83 0.96 0.91 13 反应动态 (慢偏移) 0.77 0.19 0.30 0.92 0.72 0.39 0.89 0.10 0.48 0.95 0.24 0.86 0.85 0.03 0.89 0.94 0.95 14 反应器冷却水阀门 (粘滞) 0.75 0.01 0.00 1 0.97 0.12 0.36 0.07 0.01 0.88 0.01 0.01 0.04 0.01 0.01 1 1 15 冷凝器冷却水阀门 (粘滞) 0.01 0.01 0.01 0.29 0.02 0.01 0.18 0.00 0.01 0.23 0 0.03 0.00 0.00 0.03 0.02 0.06 16 未知 0.03 0.02 0.01 0.37 0.03 0.01 0.26 0.00 0.00 0.40 0 0.85 0.03 0.01 0.05 0.15 0.11 17 未知 0.64 0.02 0.01 0.95 0.94 0.44 0.35 0.04 0.02 0.76 0.00 0.22 0.03 0.01 0.03 0.84 0.85 18 未知 0.88 0.82 0.83 0.92 0.87 0.79 0.91 0.12 0.87 0.90 0.75 0.88 0.80 0.71 0.85 0.88 0.88 19 未知 0.01 0.02 0.01 0.24 0.08 0.01 0.14 0.01 0.01 0.16 0 0.12 0.01 0.32 0.66 0.01 0.03 20 未知 0.03 0.02 0.01 0.61 0.02 0.01 0.38 0.05 0.39 0.81 0.00 0.23 0.47 0.01 0.89 0.32 0.43 21 流 4 的阀门固定在稳态位置 0.01 0.00 0.00 0.66 0.44 0.01 0.87 0.00 0.01 0.73 0.00 0.45 0.31 0.00 0.02 0.41 0.84
下载: 导出CSV
表 3 甘油精馏过程中的监测变量
Table 3 Measured variables in the distillation process
单元 1 变量名称 单元 2 变量名称 1 进料流量 1 进料储罐液位 2 灵敏板温度 2~13 塔板温度1~12 3 塔底液位 14 冷却水流量 4 塔顶回流 15 重相储灌液位 5 塔顶产品流 16 轻相储罐液位
下载: 导出CSV
-
[1] 柴天佑, 丁进良. 流程工业智能优化制造. 中国工程科学, 2018, 20(4): 51−58Chai Tian-You, Ding Jin-Liang. Smart and optimal manufacturing for process industry. Strategic Study of CAE, 2018, 20(4): 51−58 [2] 刘强, 卓洁, 郎自强, 秦泗钊. 数据驱动的工业过程运行监控与自优化研究展望. 自动化学报, 2018, 44(11): 1944−1956Liu Qiang, Zhuo Jie, Lang Zi-Qiang, Qin S Joe. Perspectives on data-driven operation monitoring and self-optimization of industrial processes. Acta Automatica Sinica, 2018, 44(11): 1944−1956 [3] 钱锋, 杜文莉, 钟伟民, 唐漾. 石油和化工行业智能优化制造若干问题及挑战. 自动化学报, 2017, 43(6): 893−901Qian Feng, Du Wen-Li, Zhong Wei-Min, Tang Yang. Problems and challenges of smart optimization manufacturing in petrochemical industries. Acta Automatica Sinica, 2017, 43(6): 893−901 [4] Qin S J. Process data analytics in the era of big data. AIChE Journal, 2014, 60(9): 3092−3100 doi: 10.1002/aic.14523 [5] 赵春晖, 余万科, 高福荣. 非平稳间歇过程数据解析与状态监控—回顾与展望, 自动化学报, 2020, DOI: 10.16383/j.aas.c190586Zhao Chun-Hui, Yu Wan-Ke, Gao Fu-Rong. Data analytics and condition monitoring methods for nonstationary batch processes—current status and future, Acta Automatica Sinica, 2020, DOI: 10.16383/j.aas.c190586 [6] Wang Y Q, Si Y B, Huang B, and Lou Z J. Survey on the theoretical research and engineering applications of multivariate statistics process monitoring algorithms: 2008–2017. The Canadian Journal of Chemical Engineering, 2018, 96(10): 2073−2085 doi: 10.1002/cjce.23249 [7] Ge Z Q. Review on data-driven modeling and monitoring for plant-wide industrial processes. Chemometrics and Intelligent Laboratory Systems, 2017, 171: 16−25 [8] Jiang Q C, Yan X F, Huang B. Review and perspectives of data-driven distributed monitoring for industrial plant-wide processes. Industrial and Engineering Chemistry Research, 2019, 58(29): 12899−12912 [9] 郭小萍, 刘诗洋, 李元. 基于稀疏残差距离的多工况过程故障检测方法研究. 自动化学报, 2019, 45(3): 617−625Guo Xiao-Ping, Liu Shi-Yang, Li Yuan. Fault detection of multi-mode processes employing sparse residual distance. Acta Automatica Sinica, 2019, 45(3): 617−625 [10] Ding S X, Data-driven Design of Fault Diagnosis and Fault-tolerant Control Systems: Springer London, 2014. [11] Zhang Y W, Zhou H, Qin S J. Decentralized fault diagnosis of large-scale processes using multiblock kernel principal component analysis. Acta Automatica Sinica, 2010, 36(4): 593−597 [12] Zhao C H, Sun H. Dynamic distributed monitoring strategy for large-scale nonstationary processes subject to frequently varying conditions under closed-loop control. IEEE Transactions on Industrial Electronics, 2019, 66(6): 4749−4758 doi: 10.1109/TIE.2018.2864703 [13] Shang C, Ji H Q, Huang X L, Yang F, Huang D X. Generalized grouped contributions for hierarchical fault diagnosis with group Lasso. Control Engineering Practice, 2019, 93: 104193 doi: 10.1016/j.conengprac.2019.104193 [14] 邹筱瑜, 王福利, 常玉清, 王敏, 蔡庆宏. 基于分层分块结构的流程工业过程运行状态评价及非优原因追溯. 自动化学报, 2019, 45(2): 315−324Zou Xiao-Yu, Wang Fu-Li, Chang Yu-Qing, Wang Min, Cai Qing-Hong. Plant-wide process operating performance assessment and non-optimal cause identication based on hierarchical multi-block structure. Acta Automatica Sinica, 2019, 45(2): 315−324 [15] Jiang Q C, Huang B. Distributed monitoring for large-scale processes based on multivariate statistical analysis and Bayesian method. Journal of Process Control, 2016, 46: 75−83 doi: 10.1016/j.jprocont.2016.08.006 [16] 罗娜, 蒋勇, 叶贞成, 杜文莉, 钱锋. 基于集成建模方法的乙二醇全流程模拟. 化工学报, 2009, 60(1): 151−156 doi: 10.3321/j.issn:0438-1157.2009.01.022Luo Na, Jiang Yong, Ye Zhen-Cheng, Du Wen-Li, Qian Feng. Simulation of ethylene glycol process based on integrated modeling method. Chinese Journal of Chemical Engineering, 2009, 60(1): 151−156 doi: 10.3321/j.issn:0438-1157.2009.01.022 [17] 文成林, 吕菲亚, 包哲静, 刘妹琴. 基于数据驱动的微小故障诊断方法综述. 自动化学报, 2016, 42(9): 1285−1299Wen Cheng-Lin, Lv Fei-Ya, Bao Zhe-Jing, Liu Mei-Qin. A review of data driven-based incipient fault diagnosis. Acta Automatica Sinica, 2016, 42(9): 1285−1299 [18] 杨浩, 姜斌, 周东华. 互联系统容错控制的研究回顾与展望. 自动化学报, 2017, 43(1): 9−19Yang Hao, Jiang Bin, Zhou Dong-Hua. Review and perspectives on fault tolerant control for interconnected systems. Acta Automatica Sinica, 2017, 43(1): 9−19 [19] Chen Z W, Ding S X, Zhang K, Li Z B, Hu Z K. Canonical correlation analysis-based fault detection methods with application to alumina evaporation process. Control Engineering Practice, 2016, 46: 51−58 doi: 10.1016/j.conengprac.2015.10.006 [20] 曹玉苹, 黄琳哲, 田学民. 一种基于 DIOCVA 的过程监控方法. 自动化学报, 2015, 41(12): 2072−2080Cao Yu-Ping, Huang Lin-Zhe, Tian Xue-Min. A process monitoring method using dynamic input-output canonical variate analysis. Acta Automatica Sinica, 2015, 41(12): 2072−2080 [21] Shi L K, Tong C D, Lan T, Shi X H. Statistical process monitoring based on ensemble structure analysis, IEEE/CAA Journal of Automatica Sinica, 2018, DOI: 10.1109/JAS. 2017.7510877 [22] 彭开香, 马亮, 张凯. 复杂工业过程质量相关的故障检测与诊断技术综述. 自动化学报, 2017, 43(3): 349−365Peng Kai-Xiang, Ma Liang, Zhang Kai. Review of quality-related fault detection and diagnosis techniques for complex industrial processes. Acta Automatica Sinica, 2017, 43(3): 349−365 [23] Ge Z Q, Song Z H. Distributed PCA model for plant-wide process monitoring. Industrial and Engineering Chemistry Research, 2013, 52(5): 1947−1957 [24] Liu Q, Qin S J, Chai T Y. Multiblock concurrent PLS for decentralized monitoring of continuous annealing processes. IEEE Transactions on Industrial Electronics, 2014, 61(61): 6429−6437 [25] Qin S J, Valle S, Piovoso M J. On unifying multiblock analysis with application to decentralized process monitoring. Journal of Chemometrics, 2001, 15(9): 715−742 doi: 10.1002/cem.667 [26] Ge Z Q. Quality prediction and analysis for large-scale processes based on multi-level principal component modeling strategy. Control Engineering Practice, 2014, 31(1): 9−23 [27] Jiang Q C, Ding S X, Wang Y, Yan X F. Data-driven distributed local fault detection for large-scale processes based on the GA-regularized canonical correlation analysis. IEEE Transactions on Industrial Electronics, 2017, 64(10): 8148−8157 doi: 10.1109/TIE.2017.2698422 [28] Wang Y, Jiang Q C, Yan X F, Fu J Q. Joint-individual monitoring of large-scale chemical processes with multiple interconnected operation units incorporating multiset CCA. Chemometrics and Intelligent Laboratory Systems, 2017, 166: 14−22 [29] Zhu J L, Ge Z Q, Song Z H. Distributed parallel PCA for modeling and monitoring of large-scale plant-wide processes with big data. IEEE Transactions on Industrial Informatics, 2017, 13(4): 1877−1885 doi: 10.1109/TII.2017.2658732 [30] Jiang Q C, Yan X F, Huang B. Performance-driven distributed PCA process monitoring based on fault-relevant variable selection and Bayesian inference. IEEE Transactions on Industrial Electronics, 2015, 63(1): 377−386 [31] Jiang B B, Huang D X, Zhu X X, Yang F, Richard D Braatz. Canonical variate analysis-based contributions for fault identification. Journal of Process Control, 2015, 26: 17−25 doi: 10.1016/j.jprocont.2014.12.001 [32] Westerhuis J A, Gurden S P, Smilde A K. Generalized contribution plots in multivariate statistical process monitoring. Chemometrics and Intelligent Laboratory Systems, 2000, 51(1): 95−114 [33] Kerkhof P, Vanlaer J, Gins G, Impe J. Analysis of smearing-out in contribution plot based fault isolation for statistical process control. Chemical Engineering Science, 2013, 104: 285−293 doi: 10.1016/j.ces.2013.08.007 [34] Yan Z B, Yao Y. Variable selection method for fault isolation using least absolute shrinkage and selection operator (LASSO). Chemometrics and Intelligent Laboratory Systems, 2015, 146: 136−146 [35] Chen Z W, Ding S X, Peng T, Yang C H, Gui W H. Fault detection for non-Gaussian processes ssing generalized canonical correlation analysis and randomized algorithms. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1559−1567 doi: 10.1109/TIE.2017.2733501 [36] Jiang Q C, Gao F R, Yi H, Yan X F. Multivariate statistical monitoring of key operation units of batch processes based on time-slice CCA. IEEE Transactions on Control Systems Technology, 2019, 27(3): 1368−1375 doi: 10.1109/TCST.2018.2803071 [37] Jiang Q C, Yan X F. Learning deep correlated representations for nonlinear process monitoring. IEEE Transactions on Industrial Informatics, 2019, 15(12): 6200−6209 doi: 10.1109/TII.2018.2886048 [38] Jiang Q C, Yan X F. Multimode process monitoring using variational Bayesian inference and canonical correlation analysis. IEEE Transactions on Automation Science and Engineering, 2019, 16(4): 1814−1824 doi: 10.1109/TASE.2019.2897477 [39] Yin S, Luo H, Ding S X. Real-time implementation of fault-tolerant control systems with performance optimization. IEEE Transactions on Industrial Electronics, 2014, 61(5): 2402−2411 doi: 10.1109/TIE.2013.2273477 [40] Downs J J, Vogel E F. A plant-wide industrial process control problem. Computers and Chemical Engineering, 1993, 17(3): 245−255 [41] Chiang L H, Russell E L, Braatz R D, Fault detection and diagnosis in industrial systems: Springer Science and Business Media, London, 2000. -
点击查看大图
计量
- 文章访问数: 1510
- HTML全文浏览量: 745
- PDF下载量: 309
- 被引次数: 0
