基于故障判别增强KECA算法的故障检测

doi:10.11949/0438-1157.20190893

摘要/Abstract

摘要：

基于核熵主成分分析方法的统计模型仅利用正常工况下数据进行建模，而忽略了监控系统数据库中一些已知类别的先前故障数据。为了利用先前故障数据中包含的故障信息来增强故障检测性能，提出了一种故障判别增强KECA（fault discriminant enhanced kernel entropy component analysis, FDKECA）算法。该法通过采用无监督学习和监督学习方法建立模型，同时监测非线性核熵主成分（kernel entropy component, KEC）和故障判别成分（fault discriminant component, FDC）两类数据特征。此外，利用贝叶斯推理将相应的监视统计信息转换为故障概率，并通过加权两个子模型的结果来构建基于总体概率的监视统计量。通过数值仿真和田纳西伊斯曼（Tennessee Eastman, TE）过程仿真实验，证明和传统KECA相比，FDKECA算法能够有效利用故障数据提高故障检测率。

关键词: 过程控制, 过程系统, 故障检测, 核熵成分分析, 混合模型

Abstract:

The statistical model based on the kernel entropy principal component analysis method only uses the data under normal operating conditions for modeling, and ignores some known categories of previous failure data in the monitoring system database. This paper proposed a fault discriminant enhanced the kernel entropy component analysis algorithm by using fault information contained in the previous fault data to enhance the fault detection performance. The model adopted unsupervised learning and supervised learning methods to monitor two types of data features: nonlinear kernel entropy component and fault discriminant component. In addition, the proposed algorithm employed Bayesian inference to convert the corresponding monitoring statistics into failure probability. The weighted probability of the two sub-models generated the monitoring statistics of the overall probability. Through numerical simulation and Tennessee Eastman process simulation experiments, the FDKECA algorithm is more effective than the traditional KECA since the improvement of the fault detection rate.

Key words: process control, process systems, fault detection, kernel entropy component analysis, mixing model

中图分类号:

TP 277

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.

Yu HAN, Junfang LI, Qiang GAO, Yu TIAN, Guogang YU. Fault detection based on fault discrimination enhanced kernel entropy component analysis algorithm[J]. CIESC Journal, 2020, 71(3): 1254-1263.

图/表 12

图1 FDKECA算法结构

Fig.1 Fault discriminant enhanced kernel entropy component analysis algorithmic structure

表1 数值仿真故障设置

Table 1 Numerical simulation fault setting

故障	故障描述	故障程度
故障	故障描述	训练集	测试集
D1	变量 $x 2$ 阶跃变化	-0.40	-0.38
D2	变量 $x 1$ 线性增加	0.01×（k-100）	0.011×（k-100）

表1 数值仿真故障设置

Table 1 Numerical simulation fault setting

故障	故障描述	故障程度
故障	故障描述	训练集	测试集
D1	变量 $x 2$ 阶跃变化	-0.40	-0.38
D2	变量 $x 1$ 线性增加	0.01×（k-100）	0.011×（k-100）

图2 KECA故障D1监控结果

Fig.2 Fault D1 monitoring results of KECA

图3 情况1训练下，FDKECA故障D1监控结果

Fig.3 FDKECA fault D1 monitoring results of case 1 training

图4 故障D1测试集投影

Fig.4 Fault D1 test projection

图5 KLNPDA故障D1监控结果

Fig.5 Fault D1 monitoring results of KLNPDA

图6 情况1下故障D1故障概率加权因子

Fig.6 Weighting factor of fault D1 for fault probability in case 1

图7 KECA故障D2监控结果

Fig.7 Fault D2 monitoring results of KECA

图8 情况1下FDKECA对故障D2监控结果

Fig.8 FDKECA fault D1 monitoring results of case 1 training

表2 三种情况下的检测率

Table 2 Fault detection rate in three cases/%

故障	KECA		FDKECA
	T²	Q	情况1		情况2		情况3
	T²	Q	PT	PQ	PT	PQ	PT	PQ
D1	15	5.5	84	7.5	45	10	85	10
D2	41	11	54.5	8.5	81.5	9	82.5	9

表3 TE过程故障检测率测试结果

Table 3 Test results of TE process fault detection rate/%

故障	KPCA		KECA		KLNPDA		FDKECA
故障	T²	Q	T²	Q	T²	Q	PT	PQ
1	99.8	99.0	99.8	99.5	99.7	99.2	100	100
2	98.4	2.6	98.4	98.8	98.6	98.5	99.8	100
3	3.2	2.8	6.4	12.6	46.2	45.5	65.2	61.8
4	98.9	84.0	95.3	93.5	96.2	91.2	99.3	98.8
5	30.8	17.0	36.5	94.3	42.7	43.7	43.6	94.3
6	99.6	50.0	99.8	100	100	99.3	100	100
7	100	77.8	100	99.9	100	99.8	100	100
8	99.0	31.4	97.6	97.7	99.5	97.6	99.3	99.0
9	3.7	3.8	17.5	19.1	64.2	45.6	89.6	52.7
10	47.8	60.8	66.7	60.8	63.7	59.6	74.0	72.5
11	69.1	66.2	57.1	67.7	74.3	67.6	75.1	72.5
12	99.0	34.8	97.5	99.2	99.7	96.5	99.7	99.7
13	95.0	12.8	92.8	95.5	95.7	95.1	95.7	95.8
14	99.9	37.1	100	99.7	100	100	100	100
15	8.9	1.1	7.1	13.1	66.2	38.7	85.1	47.3
16	31.1	54.1	58.5	51.0	62.7	40.8	74.5	63.3
17	93.5	33.2	75.1	80.1	90.1	89.5	98.6	93.8
18	90.0	2.3	91.1	92.1	91.8	92.0	93.0	93.3
19	6.0	13.4	17.6	19.8	20.0	14.6	28.3	28.8
20	58.4	47.6	59.1	58.6	63.5	69.0	68.0	73.2
21	40.0	43.5	48.0	45.3	57.0	51.6	57.7	54.2
Avg	65.3	36.9	67.6	71.3	77.7	73.1	83.1	81.0

图9 故障3监控结果

Fig.9 Monitoring results of fault 3

参考文献 33

1	Alcala C F,Qin S J.Analysis and generalization of fault diagnosis methods for process monitoring[J].Journal of Process Control,2011,21(3):322-330.
2	Ge Z,Song Z,Gao F.Review of recent research on data-based process monitoring[J].Industrial & Engineering Chemistry Research,2013,52(10):3543-3562.
3	Gao Z,Cecati C,Ding S X.A survey of fault diagnosis and fault-tolerant techniques(I): Fault diagnosis with model-based and signal-based approaches[J].IEEE Transactions on Industrial Electronics,2015,62(6):3757-3767.
4	Agrawal V,Panigrahi B K,Subbarao P M V.Review of control and fault diagnosis methods applied to coal mills[J].Journal of Process Control,2015,32:138-153.
5	Liu Y,Pan Y,Wang Q,et al.Statistical process monitoring with integration of data projection and one-class classification[J].Chemometrics and Intelligent Laboratory Systems,2015,149:1-11.
6	Jenssen R.Kernel entropy component analysis[J].IEEE Trans Pattern Anal Mach Intell,2010,32(5):847-860.
7	常鹏,乔俊飞,王普,等.基于MKECA的非高斯性和非线性共存的间歇过程监测[J].化工学报,2018,69(3):1200-1206.
	Chang P,Qiao J F,Wang P,et al.Monitoring non-Gaussian and non-linear batch process based on multi-way kernel entropy component analysis[J].CIESC Journal,2018,69(3):1200-1206.
8	齐咏生,张海利,王林,等.基于MSPCA-KECA的冷水机组故障监测及诊断[J].化工学报,2017,68(4):1499-1508.
	Qi Y S,Zhang H L,Wang L,et al.Fault detection and diagnosis for chillers using MSPCA-KECA[J].CIESC Journal,2017,68(4):1499-1508.
9	He X B,Yang Y P,Yang Y H.Fault diagnosis based on variable-weighted kernel Fisher discriminant analysis[J].Chemometrics and Intelligent Laboratory Systems,2008,93(1):27-33.
10	Zhong S,Wen Q,Ge Z.Semi-supervised Fisher discriminant analysis model for fault classification in industrial processes[J].Chemometrics and Intelligent Laboratory Systems,2014,138:203-211.
11	Zhu Z B,Song Z H.A novel fault diagnosis system using pattern classification on kernel FDA subspace[J].Expert Systems with Applications,2011,38(6):6895-6905.
12	Feng J,Wang J,Zhang H,et al.Fault diagnosis method of joint fisher discriminant analysis based on the local and global manifold learning and its kernel version[J].IEEE Transactions on Automation Science and Engineering,2016,13(1):122-133.
13	Jie Y.Nonlinear bioprocess monitoring using multiway kernel localized fisher discriminant analysis[J].Industrial & Engineering Chemistry Research,2011,50(6):3390-3402.
14	Rong G,Liu S Y,Shao J D.Fault diagnosis by locality preserving discriminant analysis and its kernel variation[J].Computers & Chemical Engineering,2013,49:105-113.
15	Yu J.Local and nonlocal preserving projection for bearing defect classification and performance assessment[J].IEEE Transactions on Industrial Electronics,2012,59(5):2363-2376.
16	Shao W,Tian X,Wang P.Supervised local and non-local structure preserving projections with application to just-in-time learning for adaptive soft sensor[J].Chinese Journal of Chemical Engineering,2015,23(12):1925-1934.
17	Yu J,Lu X.Wafer map defect detection and recognition using joint local and nonlocal linear discriminant analysis[J].IEEE Transactions on Semiconductor Manufacturing,2016,29(1):33-43.
18	Tishon A,Oldstone M B.Perturbation of differentiated functions during viral infectionin vivo.In vivo relationship of host genes and lymphocytic choriomeningitis virus to growth hormone deficiency[J].Virology,1990,142(1):158-174.
19	秦家祥.基于KECA的非线性故障检测[D].杭州:浙江大学,2017.
	Qin J X.Nonlinear fault detection based on KECA[D].Hangzhou:Zhejiang University,2017.
20	He X,Yan S,Hu Y,et al.Face recognition using laplacian faces[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(3):328-340.
21	Yang J,Zhang D,Yang J Y.Non-locality preserving projection and its application to palmprint recognition[C]//The 9th IEEE International Conference on Control,Automation, Robotics and Vision,2007:1-4.
22	Lee J M,Yoo C K,Choi S W,et al.Nonlinear process monitoring using kernel principal component analysis[J].Chemical Engineering Science,2004,59(1):223-234.
23	Lee J M,Yoo C K,Lee I B.Fault detection of batch processes using multiway kernel principal component analysis[J].Computers & Chemical Engineering,2004,28(9):1837-1847.
24	Deng X,Tian X,Chen S.Modified kernel principal component analysis based on local structure analysis and its application to nonlinear process fault diagnosis[J].Chemometrics and Intelligent Laboratory Systems,2013,127:195-209.
25	Hong X,Chen S,Harris C J.A forward-constrained regression algorithm for sparse kernel density estimation[J].IEEE Transactions on Neural Networks,2008,19(1):193-198.
26	Hong X,Gao J,Chen S,et al.Sparse density estimation on the multinomial manifold[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(11):2972-2977.
27	Deng X,Tian X,Chen S,et al.Fault discriminant enhanced kernel principal component analysis incorporating prior fault information for monitoring nonlinear processes[J].Chemometrics and Intelligent Laboratory Systems,2017,162:21-34.
28	Zhang H,Qi Y,Wang L,et al.Fault detection and diagnosis of chemical process using enhanced KECA[J].Chemometrics and Intelligent Laboratory Systems,2017,161:61-69.
29	Bernal-de-Lázaro J M,Llanes-Santiago O,Prieto-Moreno A,et al.Enhanced dynamic approach to improve the detection of small-magnitude faults[J].Chemical Engineering Science,2016,14:166-179.
30	薄翠梅,韩晓春,易辉,等.基于聚类选择 k近邻的 LLE算法及故障检测[J].化工学报,2016,67(3):925-930.
	Bo C M,Han X C,Yi H,et al.Neighborhood selection of LLEbased on cluster for fault detection[J].CIESC Journal,2016,67(3):925-930.
31	Downs J J,Vogel E F.A plant-wide industrial process control problem[J].Computers & Chemical Engineering,1993,17(3):245-255.
32	Ye N,Mcavoy T J,Kosanovich K A,et al.Plant-wide control using an inferential approach[C]//American Control Conference.IEEE,2009.
33	Huang J,Yan X.Relevant and independent multi-block approach for plant-wide process and quality-related monitoring based on KPCA and SVDD[J].ISA Transactions,2018,73:257-267.

[1]	陈哲文, 魏俊杰, 张玉明. 超临界水煤气化耦合SOFC发电系统集成及其能量转化机制[J]. 化工学报, 2023, 74(9): 3888-3902.
[2]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[3]	康飞, 吕伟光, 巨锋, 孙峙. 废锂离子电池放电路径与评价研究[J]. 化工学报, 2023, 74(9): 3903-3911.
[4]	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.
[5]	郑玉圆, 葛志伟, 韩翔宇, 王亮, 陈海生. 中高温钙基材料热化学储热的研究进展与展望[J]. 化工学报, 2023, 74(8): 3171-3192.
[6]	李贵贤, 曹阿波, 孟文亮, 王东亮, 杨勇, 周怀荣. 耦合固体氧化物电解槽的CO₂制甲醇过程设计与评价研究[J]. 化工学报, 2023, 74(7): 2999-3009.
[7]	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.
[8]	苏晓丹, 朱干宇, 李会泉, 郑光明, 孟子衡, 李防, 杨云瑞, 习本军, 崔玉. 湿法磷酸半水工艺考察与石膏结晶过程研究[J]. 化工学报, 2023, 74(4): 1805-1817.
[9]	贠程, 王倩琳, 陈锋, 张鑫, 窦站, 颜廷俊. 基于社团结构的化工过程风险演化路径深度挖掘[J]. 化工学报, 2023, 74(4): 1639-1650.
[10]	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.
[11]	张中秋, 李宏光, 石逸林. 基于人工预测调控策略的复杂化工过程多任务学习方法[J]. 化工学报, 2023, 74(3): 1195-1204.
[12]	张江淮, 赵众. 碳三加氢装置鲁棒最小协方差约束控制及应用[J]. 化工学报, 2023, 74(3): 1216-1227.
[13]	王子宗, 索寒生, 赵学良. 数字孪生智能乙烯工厂研究与构建[J]. 化工学报, 2023, 74(3): 1175-1186.
[14]	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.
[15]	王雅琳, 潘雨晴, 刘晨亮. 基于GSA-LSTM动态结构特征提取的间歇过程监测方法[J]. 化工学报, 2022, 73(9): 3994-4002.