多SVDD模型的多模态过程监控方法

doi:10.11949/j.issn.0438-1157.20150479

化工学报 ›› 2015, Vol. 66 ›› Issue (11): 4526-4533.DOI: 10.11949/j.issn.0438-1157.20150479

多SVDD模型的多模态过程监控方法

杨雅伟, 宋冰, 侍洪波

华东理工大学化工过程先进控制和优化技术教育部重点实验室, 上海 200237

收稿日期:2015-04-14 修回日期:2015-07-24 出版日期:2015-11-05 发布日期:2015-11-05
通讯作者: 侍洪波
基金资助:
国家自然科学基金项目(61374140)。

Multimode processes monitoring method via multiple SVDD model

YANG Yawei, SONG Bing, SHI Hongbo

Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education, East ChinaUniversity of Science and Technology, Shanghai 200237, China

Received:2015-04-14 Revised:2015-07-24 Online:2015-11-05 Published:2015-11-05
Supported by:
supported by the National Natural Science Foundation of China (61374140).

摘要/Abstract

摘要：

现代工业过程往往具有多个运行模态,并且单一模态中的变量服从高斯与非高斯混合的复杂数据分布。针对多模态与复杂数据分布问题,基于局部离群概率(local outlier probability, LOOP)算法与支持向量数据描述(support vector data description, SVDD)算法,提出了一种名为MSVDD(multiple support vector data description, MSVDD)的多模态过程监控方法。首先,考虑到不同模态之间存在差异,利用差分策略以及局部离群概率算法对多模态数据进行聚类。其次,在每个单一模态下分别建立SVDD模型。然后,通过计算测试样本对每个单一模态的离群概率选择合适的模型进行过程监控。最后,在Tennessee Eastman(TE)平台上进行仿真测试以验证提出方法的可行性与有效性。

关键词: 多模态, 复杂数据分布, 局部离群概率, 支持向量数据描述, 过程监控

Abstract:

Modern industrial processes always have multiple operation modes. Besides, the variable in the single mode often obey complex data distribution which is a mix of Gaussian distribution and non-Gaussian distribution. Considering the problems of both multimode and complex data distribution, a new multimode processes monitoring method called multiple SVDD is proposed based on the local outlier probability algorithm and the support vector data description algorithm. First, given that the differences exist between different modes, the clustering is conducted by employing the differential strategy and the local outlier probability algorithm. Second, the SVDD algorithm is used to build the monitoring model in each single mode. And then, the most suitable model is selected for each testing sample through calculating the outlier probability. Finally, the feasibility and efficiency are proved through the Tennessee Eastman process simulation.

Key words: multimode, complex data distribution, local outlier probability, SVDD, processes monitoring

中图分类号:

TP277

杨雅伟, 宋冰, 侍洪波. 多SVDD模型的多模态过程监控方法[J]. 化工学报, 2015, 66(11): 4526-4533.

YANG Yawei, SONG Bing, SHI Hongbo. Multimode processes monitoring method via multiple SVDD model[J]. CIESC Journal, 2015, 66(11): 4526-4533.

参考文献 22

[1]	Kruger U, Dimitriadis G. Diagnosis of process faults in chemical systems using a local partial least squares approach [J]. AIChE J., 2008, 54: 2581-2596.
[2]	Lee J M, Yoo C K, Choi S W, Vanrolleghem P A, Lee I B. Nonlinear process monitoring using kernel principal component analysis [J]. Chem. Eng. Sci., 2004, 59: 223-234.
[3]	Ge Z Q, Song Z H. Mixture Bayesian regularization method of PPCA for multi-mode process monitoring [J]. AIChE J., 2010, 56: 2838-2849.
[4]	Jin H D, Lee Y H, Han C H. Robust recursive principle component analysis modeling for adapting monitoring [J]. Ind. Eng. Chem. Res., 2006, 45: 696-703.
[5]	Zhao S J, Zhang J, Xu Y M. Monitoring of processes with multiple operation modes through multiple principle component analysis models [J]. Ind. Eng. Chem. Res., 2004, 43: 7025-7035.
[6]	Ma H H, Hu Y, Shi H B. Fault detection and identification based on the neighborhood standardized local outlier factor method [J]. Ind. Eng. Chem. Res., 2013, 52: 2389-2402.
[7]	Tan S, Wang F L, Peng J, Chang Y Q, Wang S. Multimode process monitoring based on mode identification [J]. Ind. Eng. Chem. Res., 2012, 51: 374-388.
[8]	Liu J, Chen D S. Fault detection and identification using modified Bayesian classification on PCA subspace [J]. Ind. Eng. Chem. Res., 2009, 48: 3059-3077.
[9]	Ge Z Q, Song Z H. Multimode process monitoring based on Bayesian method [J]. Journal of Chemometrics, 2009, 23: 636-650.
[10]	Xie X, Shi H B. Multimode process monitoring based on fuzzy C-means in locality preserving projection subspace [J]. Chinese J. Chem. Eng., 2012, 20:1174-1179.
[11]	Ge Z Q, Gao F R, Song Z H. Two-dimensional Bayesian monitoring method for nonlinear multimode processes [J]. Chem. Eng. Sci., 2011, 66: 5173-5183.
[12]	Choi S W, Park J H, Lee I B. Process monitoring using a Gaussian mixture model via principal component analysis and discriminant analysis [J]. Comput. Chem. Eng., 2004, 28: 1377-1387.
[13]	Yu J, Qin S J. Multimode process monitoring with Bayesian inference-based finite Gaussian mixture models [J]. AIChE J., 2008, 54:1811-1829.
[14]	Song B, Shi H B, Ma Y X, Wang J P. Multi-subspace principal component analysis with local outlier factor for multimode process monitoring [J]. Ind. Eng. Chem. Res., 2014, 53: 16453-16464.
[15]	Ge Z Q, Song Z H. Process monitoring based on independent component analysis-principal component analysis (ICA-PCA) and similarity factors [J]. Ind. Eng. Chem. Res., 2007, 46: 2054-2063.
[16]	Zhao C H, Gao F R, Wang F L. Nonlinear batch process monitoring using phase-based kernel-independent component analysis-principal component analysis (KICA-PCA) [J]. Ind. Eng. Chem. Res., 2009, 48: 9163-9174.
[17]	Ma Y X, Shi H B, Ma H H, Wang M L. Dynamic process monitoring using adaptive local outlier factor [J]. Chem. Intel. Lab. Syst., 2013, 127: 89-101.
[18]	Ge Z Q, Song Z H. Bagging support vector data description model for batch process monitoring [J]. J. Process Control, 2013, 23: 1090-1096.
[19]	Downs J J, Vogel E F. A plant-wide industrial process control problem [J]. Computers and Chemical Engineering, 1993, 17: 245-255.
[20]	Ricker N L. Optimal steady-state operation of the Tennessee Eastman challenge process [J]. Comput. Chem. Eng., 1995, 19: 949-959.
[21]	Wang L, Shi H B. Multivariate statistical process monitoring using an improved independent component analysis [J]. Chem. Eng. Res. Des., 2010, 88: 403-414.
[22]	Zhao C H, Gao F R. Fault-relevant principal component analysis (FPCA) method for multivariate statistical modeling and process monitoring [J]. Chemom. Intell. Lab. Syst., 2014, 133: 1-16.

多SVDD模型的多模态过程监控方法

Multimode processes monitoring method via multiple SVDD model

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献 22

相关文章 15

编辑推荐

Metrics

本文评价

[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]	毕荣山, 韩智慧, 陶少辉, 孙晓岩, 项曙光. 基于热扩散核密度确定密度峰值法的历史工况识别[J]. 化工学报, 2022, 73(4): 1615-1622.
[3]	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.
[4]	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.
[5]	王建林, 马琳钰, 刘伟旻, 邱科鹏, 于涛. 基于核相似度支持向量数据描述的间歇过程监测[J]. 化工学报, 2017, 68(9): 3494-3500.
[6]	郑文静, 李绍军, 蒋达. D-vine copulas混合模型及其在故障检测中的应用[J]. 化工学报, 2017, 68(7): 2851-2858.
[7]	曹玉苹, 卢霄, 田学民, 邓晓刚. 基于动态单类随机森林的非线性过程监控方法[J]. 化工学报, 2017, 68(4): 1459-1465.
[8]	顾笑伟, 王智伟, 王占伟, 何所畏, 闫增峰. 基于密度权重支持向量数据描述的冷水机组故障检测[J]. 化工学报, 2017, 68(3): 1099-1108.
[9]	刘伟旻, 王建林, 邱科鹏, 熊欢, 韩锐. 基于DHSC的多模态间歇过程测量数据异常检测方法[J]. 化工学报, 2017, 68(11): 4201-4207.
[10]	林原灵, 陈前. 基于共同趋势模型的非平稳过程在线监控[J]. 化工学报, 2017, 68(1): 178-187.
[11]	郭金玉, 袁堂明, 李元. 一种不等长的多模态间歇过程故障检测方法[J]. 化工学报, 2016, 67(7): 2916-2924.
[12]	朱红林, 王帆, 侍洪波, 谭帅. 基于非负矩阵分解的多模态过程故障监测方法[J]. 化工学报, 2016, 67(5): 1973-1981.
[13]	张汉元, 田学民. 基于KSFDA-SVDD的非线性过程故障检测方法[J]. 化工学报, 2016, 67(3): 827-832.
[14]	郑鑫, 田学民, 张汉元. 基于动态稀疏保局投影的故障检测方法[J]. 化工学报, 2016, 67(3): 833-838.
[15]	王帆, 杨雅伟, 谭帅, 侍洪波. 基于稀疏性非负矩阵分解的故障监测方法[J]. 化工学报, 2015, 66(5): 1798-1805.