化工学报 ›› 2015, Vol. 66 ›› Issue (1): 265-271.doi: 10.11949/j.issn.0438-1157.20141476

• 过程系统工程 • 上一篇    下一篇

基于统计量模式分析的T-KPLS间歇过程故障监控

常鹏, 王普, 高学金   

  1. 北京工业大学电子信息与控制工程学院, 北京 100124
  • 收稿日期:2014-09-28 修回日期:2014-10-09 出版日期:2015-01-05 发布日期:2015-01-05
  • 通讯作者: 常鹏 E-mail:changpeng2011@emails.bjut.edu.cn
  • 基金资助:

    国家自然科学基金项目(61174109,61364009)。

Fault monitoring batch process based on statistics pattern analysis of T-KPLS

CHANG Peng, WANG Pu, GAO Xuejin   

  1. College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
  • Received:2014-09-28 Revised:2014-10-09 Online:2015-01-05 Published:2015-01-05
  • Supported by:

    supported by the National Natural Science Foundation of China (61174109,61364009).

摘要:

核函数的全影结构投影(total kernel projection to latent structures,T-KPLS)最近在故障监控领域取得了广泛应用, 其实质是对数据矩阵的协方差矩阵进行分解, 没有利用数据的高阶统计量等有用信息, 在进行特征提取时会造成数据有用信息的丢失, 导致故障识别效果差。为了解决此问题, 提出了统计量模式分析(statistics pattern analysis, SPA)与核函数的全影结构投影法(total kernel projection to latent structures, T-KPLS)相结合的多向统计量模式分析的核函数的全影结构投影法(multi-way statistics pattern analysis total kernel projection to latent structures, MSPAT-KPLS)。该方法首先构造样本的不同阶次统计量, 将数据从原始的数据空间映射到统计量样本空间, 然后利用核函数将统计量样本空间映射到高维核空间并在质量变量的引导下将特征空间分为过程变量与质量变量相关、过程变量与质量变量无关、过程变量与质量变量正交和残差4个子空间;最后针对与质量变量相关和残差空间建立联合监控模型, 当监控到有故障发生时进行故障变量追溯。最后将该方法应用到微生物发酵过程中, 并与传统方法进行比较, 发现该方法具有更好的监控性能。

关键词: 故障监控, 核函数全影结构投影, 统计量模式分析

Abstract:

Total kernel projection to latent structures (T-KPLS) has been widely used in the fault detection control field, its core idea is to conduct the covariance matrix decomposition of the data matrix, without using the higher-order statistics and other useful information of the data, which will cause an information loss in the feature extraction process, then result in a bad fault recognition performance. Aiming to solve the problem, a statistics pattern analysis (SPA) combing with the T-KPLS based multi-way statistics pattern analysis total kernel projection to latent structures (MSPAT-KPLS) is proposed. First, different order statistics of the data samples are constructed to map the data from the original data space into the statistic sample space, then utilize kernel function to map the statistic sample space into the higher dimensional kernel space, and according to the quality variable, the feature space will be divided into 4 subspaces, namely: process variable related to quality variable space, process variable not related to quality variable space, process variable orthogonal to quality variable space and residual error space; Lastly, aiming at the process variable related to quality variable subspace and the residual error space, different detection models are constructed, which will trace the fault variables when faults are detected. In the end, apply the proposed method on the microbial fermentation process, and the comparison results with the traditional methods show that the proposed method could achieve a better detection.

Key words: fault monitoring, total kernel projection to latent structrues, statistics pattern analysis

中图分类号: 

  • TP273
[1] Mori J C, Yu J. Quality relevant nonlinear batch process performance monitoring using a kernel based multiway non-Gaussian latent subspace projection approach [J]. Journal of Process Control, 2014, 24(1): 57-71
[2] Xiong H S, Gong X C, Qu H B. Monitoring batch to batch reproducibility of liquid-liquid extraction process using in-line near-infrared spectroscopy combined with multivariate analysis [J]. Journal of Pharmaceutical and Biomedical Analysis, 2012, 70(11): 178-187
[3] Jia Runda (贾润达), Mao Zhizhong (毛志忠), Wang Fuli (王福利). KPLS model based product quality control for batch processes [J]. CIESC Journal ( 化工学报), 2013, 64(4): 1332-1339
[4] Stubbs S, Zhang J, Morris J L. Multiway interval partial least squares for batch process performance monitoring [J]. Industrial and Engineering Chemistry Research, 2014, 52 (35) : 12399-12407.
[5] Geert G, Jef V, Jan F. Discriminating between critical and noncritical disturbances in (bio)chemical bach processes using multi-model fault detection and end-quality prediction [J]. Industrial and Engineering Chemistry Research, 2012, 51(1): 12375-1238
[6] Naes T, Tomic O. Multi-block regression based on combination so for thogonalisation, PLS regressionand canonical correlation analysis [J]. Chemometrics and Intelligent Laboratory Systems, 2013, 124: 32-42
[7] Li G, Alcala C F, Qin S J, et al. Output relevant fault reconstruction and fault subspace extraction in total projection to latent structures models [J]. Industrial and Engineering Chemistry Research, 2010, 49(19):9175-9183
[8] Zhou D, Li G, Qin S J. Total projection to latent structures for process monitoring [J]. American Institute of Chemical Engineers, 2010, 56(1): 168-178
[9] Li G, Alcala C F, Qin S J, et al. Generalized reconstruction based contributions for output relevant fault diagnosis with application to the Tennessee Eastman process [J]. IEEE Transaction on Control Systems Technology, 2011, 19(5): 1114-1127
[10] Jackson J E. A User's Guide to Principal Components[M]. New York: Wiley, 1991
[11] Nomikos P, MacGregor J F. Monitoring batch process using multiway principal component analysis [J]. American Institute of Chemical Engineers, 1994, 40(8): 1361-1375
[12] Peng K X, Zhang K, Li G, et al. Contribution rate plot for nonlinear quality related fault diagnosis with application to the hot strip mill process [J]. Control Engineering Practice, 2013, 21(4): 360-369
[13] He Q, Wang J. Statistics pattern analysis: a new process monitoring framework and its application to semiconductor batch processes [J]. American Institute of Chemical Engineers, 2011, 57(1): 107-121
[14] Wang J, He Q. Multivariate statistical process monitoring based on statistics pattern analysis [J]. Industrial Engineering Chemical Research, 2010, 49(1): 7858-7869
[15] Zhang Hanyuan(张汉元), Tian Xuemin(田学民), Deng Xiaogang(邓晓刚). Fault identification method based on SPA similarity factor [J]. CIESC Journal(化工学报), 2013, 64(12): 4503-4508
[16] Chang Peng(常鹏),Wang Pu(王普),Gao Xuejin(高学金),et al. Batch process monitroing and quality prediction based on statistics pattern analysis and MKPLS [J]. Chinese Journal of Scientific Instrument(仪器仪表学报), 2014, 35(6): 1409-1416
[17] Birol G, Undey C, Cinar A. A modular simulation package for fed-batch fermentation: penicillin production [J]. Computers and Chemical Engineering, 2002, 26(11): 1553-1565
[18] Agudo D, Ferrer A, Ferrer J, et al. Multivariate SPC of a sequencing batch reactor for wastewater treatment [J]. Chemometrics and Intelligent Laboratory Systems, 2007, 85(1): 82-93
[19] Alcala C F, Qin S J. Analysis and generalization of fault diagnosis methods for process monitoring [J]. Journal of Process Control, 2011, 21(1): 322-330
[20] Alcala C F, Qin S J. Reconstruction based contribution for process monitoring [J]. Automatica, 2009, 45(1): 1593-1600
[1] 孙栓柱, 董顺, 江叶峰, 周挺, 李益国. 基于最小充分统计量模式分析的故障检测方法[J]. 化工学报, 2018, 69(3): 1228-1237.
[2] 张汉元, 田学民, 邓晓刚. 基于SPA相似系数的故障识别方法[J]. 化工学报, 2013, 64(12): 4503-4508.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 刘植昌, 孟祥海, 徐春明, 高金森. 重油催化裂解汽柴油二次裂解性能研究[J]. CIESC Journal, 2007, 15(3): 309 -314 .
[2] 刘永健, 袁希钢, 罗祎青. 基于浓度间隔分析的用水系统集成(II)不连续过程[J]. CIESC Journal, 2007, 15(3): 369 -375 .
[3] 刘亚青, 赵贵哲. 三聚氯化磷腈微胶囊阻燃剂/聚丙烯复合材料的性能研究[J]. CIESC Journal, 2007, 15(3): 429 -432 .
[4] 高习群, 马友光, 朱春英, 余国琮. 单泡吸收过程的界面传质机理[J]. CIESC Journal, 2006, 14(2): 158 -163 .
[5] 王淑梅, 于养信, 高光华. 气体透过碳膜的非平衡动力学模拟研究
[J]. CIESC Journal, 2006, 14(2): 164 -170 .
[6] 马永锡, 张红. 基于中心复合设计的振荡热管传热性能分析
[J]. CIESC Journal, 2006, 14(2): 223 -228 .
[7] 周笑鹏, 史清洪, 邢新会, 孙彦. 快速纯化在大肠杆菌中表达的增强型绿色荧光蛋白
[J]. CIESC Journal, 2006, 14(2): 229 -234 .
[8] 毛立新, 高翔, 张猛响, 金日光. 两步共混工艺制备聚碳酸酯/聚丙烯/针状硅酸盐三元纳米复合材料:降解与形态[J]. CIESC Journal, 2006, 14(2): 248 -252 .
[9] 张颖, 周荣琪. 常压下H2O+HCl+MgCl2与H2O+HCl+CaCl2体系的汽液平衡[J]. CIESC Journal, 2006, 14(2): 276 -280 .
[10] 王小芳, 金保升, 熊源泉, 钟文琪. Flow Behaviors of Gas-Solid Injector by 3D Simulation with Kinetic Theory of Granular Flow[J]. Chinese Journal of Chemical Engineering, 2008, 16(6): 823 -831 .