基于变量分组DTW-MCVA的不等长间歇过程故障检测方法

doi:10.11949/0438-1157.20190349

摘要/Abstract

摘要：

针对不等长间歇过程监控中批次数据同步化未能充分挖掘局部信息的问题，提出一种基于变量分组DTW-MCVA（VGDTW-CVA）的不等长间歇过程故障检测方法。首先，利用互信息矩阵描述不等长间歇过程测量变量之间的相关性，并基于互信息矩阵进行变量分组。然后利用DTW算法对各个变量组分别进行同步化，并将同步化后的变量组整合为完整的三维数据集。最后，利用MCVA方法建立动态监控模型实现对间歇生产过程的在线监控。盘尼西林发酵过程的仿真结果表明，VGDTW-MCVA能够比基本的DTW-MCVA方法更好地监控间歇过程故障。

关键词: 不等长间歇过程, 局部信息挖掘, 变量分组, 互信息, 动态时间规整, 多向典型变量分析, 故障检测

Abstract:

Aiming at the problem that batch data synchronization in unequal-length batch process monitoring fails to fully exploit local information, this paper proposes a variable grouping dynamic time warping-multiway canonical variate analysis (VGDTW-CVA) algorithm for unequal-length batch process fault detection. First, the mutual information matrix is used to describe the correlation between variables of unequal-length batch process, and all the variables are divided into serval groups based on mutual information matrix. Then use the DTW algorithm to synchronize each variable group separately, and integrate the synchronized variable groups into a complete 3D data set. Finally, MCVA method is utilized to establish dynamic monitoring model for online monitoring of batch production process. The simulation results on the penicillin simulation system show that the VGDTW-MCVA has better monitoring effect for the unequal-length batch production process.

Key words: unequal-length batch process, local information exploiting, variable grouping, mutual information, dynamic time warping, canonical variate analysis, fault detection

中图分类号:

TP 277

于蕾, 邓晓刚, 曹玉苹, 路凯琪. 基于变量分组DTW-MCVA的不等长间歇过程故障检测方法[J]. 化工学报, 2019, 70(9): 3441-3448.

Lei YU, Xiaogang DENG, Yuping CAO, Kaiqi LU. Fault detection method of unequal-length batch process based on VGDTW-MCVA[J]. CIESC Journal, 2019, 70(9): 3441-3448.

图/表 10

图1 DTW算法原理示意图

Fig.1 Schematic diagram of DTW algorithm

图2 批次-变量展开方式

Fig.2 Batch-variable expansion mode

图3 Pensim发酵过程流程图

Fig.3 Flow diagram of Pensim fermentation process

表1 故障描述

Table 1 Fault description

故障	描述
1	底物流加速率在100~400 h施加+0.004/100 h斜坡干扰
2	通风速率在100~400 h施加+0.5/100 h斜坡干扰
3	搅拌速率在100~400 h施加+3/100 h斜坡干扰

表2 变量分组

Table 2 Variable Grouping

序号	变量号
1	1、2、3
2	4、7、8
3	5、6、9、10

图4 故障1监控结果

Fig.4 Monitoring result of fault 1

表3 3种故障的检出率

Table 3 Detection rate of 3 faults/%

故障	DTW-MCVA		VGDTW-MPCA		VGDTW-MCVA
故障	T ²	SPE	T ²	SPE	T ²	SPE
1	67.79	61.07	64.00	70.00	75.84	80.54
2	94.63	85.91	88.00	78.67	95.97	91.95
3	94.63	93.96	80.67	80.00	83.22	97.32

表4 3种故障的检出采样点

Table 4 Detection sample of 3 faults

故障	DTW-MCVA		VGDTW-MPCA		VGDTW-MCVA
故障	T ²	SPE	T ²	SPE	T ²	SPE
1	101	111	105	96	89	84
2	61	78	69	80	61	65
3	61	62	80	81	83	57

图5 故障2监控结果

Fig.5 Monitoring result of fault 2

图6 故障3监控结果

Fig.6 Monitoring result of fault 3

参考文献 31

1	Ge Z Q , Song Z H , Gao F R . Review of recent research on data-based process monitoring[J]. Industrial & Engineering Chemistry Research, 2013, 52(10): 3543-3562.
2	Yang C M , Hou J . Fed-batch fermentation penicillin process fault diagnosis and detection based on support vector machine[J]. Neurocomputing, 2016, 190(2): 117-123.
3	高学金, 黄梦丹, 王普, 等 . 基于多约束DTW的MPCA间歇过程监测方法[J]. 北京工业大学学报, 2018, 44(3): 393-400.
	Gao X J , Huang M D , Wang P , et al . Batch process monitoring method based on multi-way principal component analysis and limited-DTW[J]. Journal of Beijing University of Technology, 2018, 44(3): 393-400.
4	Yin S , Ding S X , Abandan Sari A H , et al . Data-driven monitoring for stochastic systems and its application on batch process[J]. International Journal of Systems Science, 2013, 44(7): 1366-1376.
5	Liu Y Q , Pan Y P , Wang Q L , et al . Statistical process monitoring with integration of data projection and one-class classification[J]. Chemometrics and Intelligent Laboratory Systems, 2015, 149(8): 1-11.
6	Naes T , Tomic O . Multi-block regression based on combinations of orthogonalisation PLS regression and canonical correlation analysis[J]. Chemometrics and Intelligent Laboratory Systems, 2013, 124(1): 32-42.
7	Zhao X Q , Wang T . Tensor dynamic neighborhood preserving embedding algorithm for fault diagnosis of batch process[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 162(1): 94-103.
8	Cai B P , Zhao Y B , Liu H L , et al . A data-driven fault diagnosis methodology in three-phase inverters for PMSM drive systems[J]. IEEE Transactions on Power Electronics, 2017, 32(7): 5590-5600.
9	Deng X G , Tian X M . Sparse kernel locality preserving projection and its application in nonlinear process fault detection[J]. Chinese Journal of Chemical Engineering, 2013, 21(2): 163-170.
10	Song Y , Jiang Q C , Yan X F . Fault diagnosis and process monitoring using a statistical pattern framework based on a self-organizing map[J]. Journal of Central South University, 2015, 22(2): 601-609.
11	Tong C , Lan T , Shi X . Ensemble modified independent component analysis for enhanced non-Gaussian process monitoring[J]. Control Engineering Practice, 2017, 58(9): 34-41.
12	Erdem O , Ceyhan E , Varli Y . A new correlation coefficient for bivariate time-series data[J]. Physica A: Statistical Mechanics and Its Applications 2011, 414(10): 274-284.
13	Kano M , Tanaka S , Hasebe S , et al . Monitoring independent components for fault detection[J]. AIChE Journal, 2010, 49(4): 969-976.
14	Huang J P , Yan X F . Relevant and independent multi-block approach for plant-wide process and quality-related monitoring based on KPCA and SVDD[J]. ISA Transactions, 2018, 73(6): 257-267.
15	王振雷, 江伟, 王昕 . 基于多块MICA-PCA的全流程过程监控方法[J]. 控制与决策, 2018, 33(2): 269-274.
	Wang Z L , Jiang W , Wang X . Plant-wide process monitoring based on multiblock MICA-PCA[J]. Control and Decision, 2018, 33(2): 269-274.
16	Lee J M , Yoo C K , Lee I B . On-line batch process monitoring using a consecutively updated multiway principal component analysis model[J]. Computers & Chemical Engineering, 2003, 27(12): 1903-1912.
17	Rato T J , Blue J , Pinaton J , et al . Translation-invariant multiscale energy-based PCA for monitoring batch processes in semiconductor manufacturing[J]. IEEE Transactions on Automation Science & Engineering, 2017, 14(2): 894-904.
18	赵忠盖, 刘飞.基于稀疏核主元分析的在线非线性过程监控[J]. 化工学报, 2008, 59(7): 1773-1777.
	Zhao Z , Liu F . On-line nonlinear process monitoring based on sparse kennel principal component analysis[J]. Journal of Chemical Industry and Engineering(China), 2008, 59(7): 1773-1777.
19	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 . Fault detection method based on variable sub-region PCA[J]. CIESC Journal, 2016, 67(10): 4300-4308.
20	汤健, 贾美英, 刘卓 . 基于偏最小二乘算法的高维谱数据特征选择[J]. 控制工程, 2015, 22(6): 1127-1130.
	Tang J , Jia M Y , Liu Z , et al . Feature selection approach of high dimension spectral data based on partial least squares algorithm[J]. Control Engineering of China, 2015, 22(6): 1127-1130.
21	Cao Y P , Hu Y P , Deng X G , et al . Quality-relevant batch process fault detection using a multiway multi-subspace CVA method[J]. IEEE Access, 2017, 5(10): 2169-3536.
22	Juricek B C , Seborg D E , Larimore W E . Fault detection using canonical variate analysis[J]. Industrial & Engineering Chemistry Research, 2004, 43(2): 458-474.
23	Liu Y , Wang F , Chang Y , et al . Multiple hypotheses testing-based operating optimality assessment and nonoptimal cause identification for multiphase uneven-length batch process[J]. Industrial & Engineering Chemistry Research, 2016, 55(2): 6133-6144.
24	Spooner M , Kold D , Kulahci M . Selecting local constraint for alignment of batch process data with dynamic time warping[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 167(8): 161-170.
25	Lu B , Xu S , Stuber J , et al . Constrained selective dynamic time warping of trajectories in three dimensional batch data[J]. Chemommetrics and Intelligent Laboratory Systems. 2016, 159(10): 138-150.
26	Gollmer K , Posten C . Detection of distorted pattern using dynamic time warping algorithm and application for supervision of bioprocess[J]. IFAC Proceedings Volumes, 1995, 28(12): 538-549.
27	高翔, 王纲, 马纪虎 . 基于可变多元统计模型的故障诊断方法[J]. 上海海事大学学报, 2001, 22(3): 51-55.
	Gao X , Wang G , Ma J H . The approach of fault diagnosis method based on variable multivariate statistical model[J]. Journal of Shanghai Maritime University, 2001, 22(3): 51-55.
28	李元, 王纲, 曹锐 . 基于MPCA与DTW的间歇反应过程的性能监视及故障诊断[J]. 沈阳化工大学学报, 2003, 17(4): 285-289.
	Li Y , Wang G , Cao R . Performance monitoring and fault diagnosis based on the intermittent reaction process of MPCA and DTW [J]. Journal of Shenyang Chemical University, 2003, 17(4): 285-289.
29	肖应旺, 徐保国 . 基于改进的MPCA与DTW方法及其在批过程故障诊断中的应用[J]. 计算机与应用化学, 2005, 22(5): 344-348.
	Xiao Y W , Xu B G . Batch process monitoring and fault detection based on an improved MPCA and DTW[J]. Computers and Applied Chemistry, 2005, 22(5): 344-348.
30	Birol G , Ündey C , Parulekar S J , et al . A morphologically structured model for penicillin production[J]. Biotechnology and Bioengineering, 2002, 77(5): 538-552.
31	刘毅, 王海清 . Pensim仿真平台在青霉素发酵过程的应用研究[J]. 系统仿真学报, 2006, 18(12): 3524-3527.
	Liu Y , Wang H Q . Pensim simulation platform in the application of penicillin fermentation process research [J]. Journal of System Simulation, 2006, 18(12): 3524-3527.

[1]	张逸豪, 王振雷. 基于最大信息系数的分组支持向量数据描述故障检测[J]. 化工学报, 2023, 74(9): 3865-3878.
[2]	邵远哲, 赵忠盖, 刘飞. 基于共同趋势模型的非平稳过程质量相关故障检测方法[J]. 化工学报, 2023, 74(6): 2522-2537.
[3]	宋冰, 郑城风, 侍洪波, 陶阳, 谭帅. 基于VAE-OCCA的质量相关故障检测方法研究[J]. 化工学报, 2023, 74(4): 1630-1638.
[4]	杨明辉, 刘晓月, 邓晓刚, 廖明燕, 侯春望. 基于加权概率CVDA的动态化工系统微小故障检测[J]. 化工学报, 2022, 73(9): 3963-3972.
[5]	雍加望, 赵倩倩, 冯能莲. 基于非线性动态模型的质子交换膜燃料电池故障诊断[J]. 化工学报, 2022, 73(9): 3983-3993.
[6]	郭金玉, 王哲, 李元. 基于核熵独立成分分析的故障检测方法[J]. 化工学报, 2022, 73(8): 3647-3658.
[7]	王琨, 侍洪波, 谭帅, 宋冰, 陶阳. 局部时差约束邻域保持嵌入算法在故障检测中的应用[J]. 化工学报, 2022, 73(7): 3109-3119.
[8]	郭金玉, 李文涛, 李元. 在线压缩KECA的自适应算法在故障检测中的应用[J]. 化工学报, 2021, 72(8): 4227-4238.
[9]	李元, 杨东昇, 赵丽颖, 张成. 层次变分高斯混合模型与主多项式分析的故障检测策略[J]. 化工学报, 2021, 72(3): 1616-1626.
[10]	王晓慧, 王延江, 邓晓刚, 张政. 基于加权深度支持向量数据描述的工业过程故障检测[J]. 化工学报, 2021, 72(11): 5707-5716.
[11]	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.
[12]	韩宇, 李俊芳, 高强, 田宇, 禹国刚. 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.
[13]	徐静,王振雷,王昕. 基于非线性动态全局局部保留投影算法的化工过程故障检测[J]. 化工学报, 2020, 71(12): 5655-5663.
[14]	孙中建,杨博,齐楚,李宏光. 面向工业混杂系统故障检测的扩展数据逻辑分析方法[J]. 化工学报, 2020, 71(11): 5237-5245.
[15]	柴伟, 郭龙航, 池彬彬. 污水处理厂出水水质变量区间预测建模[J]. 化工学报, 2019, 70(9): 3449-3457.