基于多层优化PCC-SDG方法的化工过程故障诊断

doi:10.11949/j.issn.0438-1157.20171104

化工学报 ›› 2018, Vol. 69 ›› Issue (3): 1173-1181.DOI: 10.11949/j.issn.0438-1157.20171104

基于多层优化PCC-SDG方法的化工过程故障诊断

董玉玺, 李乐宁, 田文德

青岛科技大学化工学院, 山东青岛 266042

收稿日期:2017-08-14 修回日期:2017-08-20 出版日期:2018-03-05 发布日期:2018-03-05
通讯作者: 田文德
基金资助:
国家自然科学基金项目（21576143）。

A novel fault diagnosis method based on multilayer optimized PCC-SDG

DONG Yuxi, LI Lening, TIAN Wende

College of Chemical Engineering, Qingdao University of Science & Technology, Qingdao 266042, Shandong, China

Received:2017-08-14 Revised:2017-08-20 Online:2018-03-05 Published:2018-03-05
Supported by:
supported by the National Natural Science Foundation of China (21576143).

摘要/Abstract

摘要：

化工过程的故障发生往往都是一个变量带动多个变量的连锁效应。本文基于变量的相关性变化特点，用符号有向图SDG（signed directed graph）描述系统因果影响关系，以皮尔逊相关系数PCC（Pearson correlation coefficient）计算网络统计指标，提出了一种基于多层优化PCC-SDG的故障诊断方法。该方法基于全工艺的网络拓扑结构，首先对选取的变量进行初步优化。然后，为有效提取工艺特征信息，运用PCA（principal component analysis）权重思想从多层相关系数集中选取了权重较大的关键变量，结合SDG建立最优PCC-SDG网络。最后，针对最优PCC-SDG网络变量的相关性规律重构聚集权重系数Q，进行过程故障检测与诊断。TE（Tennessee Eastman）仿真过程的应用结果表明，PCC-SDG建模及故障诊断步骤较为简洁，可以充分挖掘SDG深层次关联特性，定量简化SDG的故障诊断效果明显，具有较好的过程监控优势。

关键词: 故障检测与诊断, 皮尔逊相关系数, SDG, 多层相关系数集, 聚集权重系数Q规则

Abstract:

Chemical process failures are often caused by a series of variables with a chain effect. This study utilizes variable correlation characteristics, PCC (Pearson correlation coefficient) statistical index, and SDG (signed directed graph) to describe the causal relationship among variables, and then proposes a PCC-SDG fault diagnosis method based on a multi-layer optimization structure. With the topological network structure of the whole process as reference, this method first performs an initial optimization on the selected variable. An optimal PCC-SDG network is then constructed on the specific variables which have large PCA (principal component analysis) weights in the multilayer correlation coefficient set. After that, the rule of gather weighting coefficient Q is established to identify process fault. The application on Tennessee Eastman process illustrates that the PCC-SDG method can realize fault detection and isolation tasks in an effective pattern. Because its modeling and diagnosis procedures are simple and SDG can be readily probed for the root cause, the proposed method has an advantage in process supervision.

Key words: fault detection and diagnosis, Pearson correlation coefficient, SDG, multilayer correlation coefficient sets, rule of gather weighting coefficient Q

中图分类号:

TQ018

董玉玺, 李乐宁, 田文德. 基于多层优化PCC-SDG方法的化工过程故障诊断[J]. 化工学报, 2018, 69(3): 1173-1181.

DONG Yuxi, LI Lening, TIAN Wende. A novel fault diagnosis method based on multilayer optimized PCC-SDG[J]. CIESC Journal, 2018, 69(3): 1173-1181.

参考文献 30

[1]	YU J, RASHID M M. A novel dynamic Bayesian network based networked process monitoring approach for fault detection, propagation identification, and root cause diagnosis[J]. AIChE Journal, 2013, 59(7):2348-2365.
[2]	耿志强, 武开英, 韩永明. 基于层次分析的FLANN神经网络研究及应用[J]. 化工学报, 2016, 67(3):805-811. GENG Z Q, WU K Y, HAN Y M. Research and application of FLANN neural network based on analytic hierarchy process[J]. CIESC Journal, 2016, 67(3):805-811.
[3]	ESLAMLOUEYAN R. Designing a hierarchical neural network based on fuzzy clustering for fault diagnosis of the Tennessee-Eastman process[J]. Applied Soft Computing, 2011, 11(1):1407-1415.
[4]	NUCHITPRASITTICHAI A, CREMASCHI S. An algorithm to determine sample sizes for optimization with artificial neural networks[J]. AIChE Journal, 2013, 59(3):805-812.
[5]	ZHANG Y W, LI S, TENG Y D. Dynamic processes monitoring using recursive kernel principal component analysis[J]. Chemical Engineering Science, 2012, 72(16):78-86.
[6]	JIANG Q C, YAN X F, ZHAO W X. Fault detection and diagnosis in chemical processes using sensitive principal component analysis[J]. Industrial & Engineering Chemistry Research, 2013, 52(4):1635-1644.
[7]	徐莹, 邓晓刚, 钟娜. 基于ICA混合模型的多工况过程故障诊断方法[J]. 化工学报, 2016, 67(9):3793-3803. XU Y, DENG X G, ZHONG N. A fault diagnosis method for multimode processes based on ICA mixture models[J]. CIESC Journal, 2016, 67(9):3793-3803.
[8]	JIANG Q C, YAN X F. Just-in-time reorganized PCA integrated with SVDD for chemical process monitoring[J]. AIChE Journal, 2014, 60(3):949-965.
[9]	YU J. Online quality prediction of nonlinear and non-Gaussian chemical processes with shifting dynamics using finite mixture model based Gaussian process regression approach[J]. Chemical Engineering Science, 2012, 82(1):22-30.
[10]	童楚东, 史旭华. 基于互信息的PCA方法及其在过程监测中的应用[J]. 化工学报, 2015, 66(10):4101-4106. TONG C D, SHI X H. Mutual information based PCA algorithm with application in process monitoring[J]. CIESC Journal, 2015, 66(10):4101-4106.
[11]	江伟, 王振雷, 王昕. 基于混合分块DMICA-PCA的全流程过程监控方法[J]. 化工学报, 2016, 67(2):759-766. JIANG W, WANG Z D, WANG X. Plant-wide process monitoring based on mixed multiblock DMICA-PCA[J]. CIESC Journal, 2016, 67(2):759-766.
[12]	ZHANG Y. Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM[J]. Chemical Engineering Science, 2009, 64(5):801-811.
[13]	潘吉铮, 周传光, 钱宇. 基于神经网络的化工过程测量数据在线校正技术的研究[J]. 高校化学工程学报, 2003, 17(3):319-324. PAN J Z, ZHOU C G, QIAN Y. On-line data rectification for chemical process measurements based on artificial neural network[J]. J. Chem. Eng. of Chinese Univ., 2003, 17(3):319-324.
[14]	王磊, 邓晓刚, 徐莹, 等. 基于变量子域PCA的故障检测方法[J]. 化工学报, 2016, 67(10):4300-4308. WANG L, DENG X G, XU Y, et al. A fault detection method based on variable sub-region[J]. CIESC Journal, 2016, 67(10):4300-4308.
[15]	衷路生, 何东, 龚锦红, 等. 基于分布式ICA-PCA模型的工业过程故障监测[J]. 化工学报, 2015, 66(11):4546-4554. ZHONG L S, HE D, GONG J H, et al. Industrial process fault monitoring based on distributed ICA-PCA model[J]. CIESC Journal, 2015, 66(11):4546-4554.
[16]	梁晴晴, 韩华, 崔晓钰, 等. 基于主元分析-概率神经网络的制冷系统故障诊断[J]. 化工学报, 2016, 67(3):1022-1031. LIANG Q Q, HAN H, CUI X Y, et al. Fault diagnosis of refrigeration system based on principal component analysis-probabilistic neural network[J]. CIESC Journal, 2016, 67(3):1022-1031.
[17]	VEDAM H, VENKATASUBRAMANIAN V. PCA-SDG based process monitoring and fault diagnosis[J]. Control Engineering Practice, 1999, 7(7):903-917.
[18]	WANG G J, XIE C, CHEN S, et al. Random matrix theory analysis of cross-correlations in the US stock market:evidence from Pearson's correlation coefficient and detrended cross-correlation coefficient[J]. Physica A:Statistical Mechanics and Its Applications, 2013, 392(17):3715-3730.
[19]	ZHOU H, DENG Z, XIA Y, et al. A new sampling method in particle filter based on Pearson correlation coefficient[J]. Neurocomputing, 2016, 216:208-215.
[20]	SALLEH F H M, ARIF S M, ZAINUDIN S, et al. Reconstructing gene regulatory networks from knock-out data using Gaussian Noise Model and Pearson Correlation Coefficient[J]. Computational Biology and Chemistry, 2015, 59:3-14.
[21]	王再英, 白华宁. 基于相关系数的过程系统故障检测与诊断方法[J]. 化工学报, 2013, 64(12):4621-4627. WANG Z Y, BAI H N. A method for fault detection and diagnosis of process system based on correlation coefficient[J]. CIESC Journal, 2013, 64(12):4621-4627.
[22]	卢秉南, 张贝克, 马昕, 等. 基于SDG模型的控制系统故障诊断方法[J]. 化工学报, 2009, 60(9):2243-2251. LU B N, ZHANG B K, MA X, et al. Fault diagnosis of control system based on SDG model[J]. CIESC Journal, 2009, 60(9):2243-2251.
[23]	FAN Y, SHAH S L, XIAO D Y. SDG (signed directed graph) based process description and fault propagation analysis for a tailings pumping process[J]. IFAC Proceedings Volumes, 2010, 43(9):50-55.
[24]	王杭州, 陈丙珍, 何小荣, 等. 基于开源组件的SDG推理平台[J]. 化工学报, 2010, 61(7):1829-1836. WANG H Z, CHEN B Z, HE X R, et al. Open source signed digraph inference framework[J]. CIESC Journal, 2010, 61(7):1829-1836.
[25]	ZHANG B K, XU X, MA X, et al. SDG-based model validation in chemical process simulation[J]. Chinese Journal of Chemical Engineering, 2013, 21(8):876-885.
[26]	GAO D, WU C G, ZHANG B K, et al. Signed directed graph and qualitative trend analysis based fault dianosis in chemical industry[J]. Chinese Journal of Chemical Engineering, 2010, 18(2):265-276.
[27]	XU W, MA R, ZHOU Y, et al. Asymptotic properties of Pearson? s rank-variate correlation coefficient in bivariate normal model[J]. Signal Processing, 2016, 119(C):190-202.
[28]	KIM Y, KIM T H, ERGUN T. The instability of the Pearson correlation coefficient in the presence of coincidencaloutliers[J]. Finance Research Letters, 2015, 13:243-257.
[29]	D'ANGELO M, PALHARES R, CAMARGOS F M, et al. A new fault classification approach applied to Tennessee Eastman benchmark process[J]. Applied Soft Computing, 2016, 49:676-686.
[30]	TIAN W D, SUN S L, GUO Q J. Fault detection and diagnosis for distillation column using two-tier model[J]. The Canadian Journal of Chemical Engineering, 2013, 91(10):1671-1685.

基于多层优化PCC-SDG方法的化工过程故障诊断

A novel fault diagnosis method based on multilayer optimized PCC-SDG

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 30

相关文章 6

编辑推荐

Metrics

本文评价

[1]	邓明月, 刘建昌, 许鹏, 谭树彬, 商亮亮. 基于KECA的非线性工业过程故障检测与诊断新方法[J]. 化工学报, 2020, 71(5): 2151-2163.
[2]	江升, 旷天亮, 李秀喜. 基于稀疏过滤特征学习的化工过程故障检测方法[J]. 化工学报, 2019, 70(12): 4698-4709.
[3]	王路瑶, 吴斌, 杜志敏, 晋欣桥. 基于长短期记忆神经网络的数据中心空调系统传感器故障诊断[J]. 化工学报, 2018, 69(S2): 252-259.
[4]	卢秉南, 张贝克, 马昕, 许欣, 高东. 基于SDG模型的控制系统故障诊断方法 [J]. 化工学报, 2009, 60(9): 2243-2251.
[5]	张贝克;郑然;马昕;吴重光. 间歇过程动态SDG建模 [J]. CIESC Journal, 2008, 59(7): 1863-1868.
[6]	陈耀,王文海,孙优贤. 基于动态主元分析的统计过程监视 [J]. CIESC Journal, 2000, 51(5): 666-670.