基于局部密度估计的多模态过程故障检测

doi:10.3969/j.issn.0438-1157.2014.08.031

化工学报 ›› 2014, Vol. 65 ›› Issue (8): 3071-3081.DOI: 10.3969/j.issn.0438-1157.2014.08.031

基于局部密度估计的多模态过程故障检测

刘帮莉, 马玉鑫, 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:2013-12-03 修回日期:2014-03-20 出版日期:2014-08-05 发布日期:2014-08-05
通讯作者: 侍洪波
基金资助:
国家自然科学基金项目（61374140）；上海浦江计划项目（12PJ1402200）。

Multimode process monitoring based on local density estimation

LIU Bangli, MA Yuxin, SHI Hongbo

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

Received:2013-12-03 Revised:2014-03-20 Online:2014-08-05 Published:2014-08-05
Supported by:
supported by the National Natural Science Foundation of China (61374140) and Shanghai Pujiang Program (12PJ1402200).

摘要/Abstract

摘要： 以市场需求为导向的现代工业过程的生产条件要根据市场的需求不断做出调整，因此实际工业过程中存在多种工况的复杂情况，而过程的数据将不再完全服从高斯分布，其均值与协方差结构往往随着工况的切换而发生较大变化，为了能及时检测此类生产过程中的故障，提出一种新的基于带宽可变的局部密度估计的过程在线监控策略。首先利用局部投影保留（locality preserving projection， LPP）将高维数据投影到低维子空间中，充分地保留数据的局部结构；然后通过带宽可变的非参数密度核函数来进行局部密度估计，并采用局部密度因子（local density factor， LDF）的思想构造监控统计量，进而对工业过程故障进行在线检测；最后通过仿真研究，结果表明所提方法能够有效地应用于多模态过程的故障检测。

关键词: 多模态过程系统, 局部投影保留算法, 带宽优化, 局部密度因子, 监控模型, 仿真实验

Abstract: Multiple operating modes and complex data distributions are common in modern industrial processes, so the process variable does not completely obey Gaussian distribution, and its mean and covariance structure tend to change with the switching of working conditions. In order to detect various faults in such production process timely, a novel online monitoring strategy based on locality preserving projection (LPP) and local density estimation with variable bandwidth is proposed. Firstly, LPP is introduced for feature extraction and dimension reduction, and then a variable bandwidth kernel density estimate function is adopted to compute every sample's local density factor, which is used as a monitoring statistic value. Finally, the results of numerical examples and TE chemical process simulation experiment validate the effectiveness and utility of the proposed method for the fault detection of multi-mode processes.

Key words: multi-mode process systems, locality preserving projection algorithm, bandwidth optimization, local density factor, monitoring model, simulation experiment

中图分类号:

TP277

刘帮莉, 马玉鑫, 侍洪波. 基于局部密度估计的多模态过程故障检测[J]. 化工学报, 2014, 65(8): 3071-3081.

LIU Bangli, MA Yuxin, SHI Hongbo. Multimode process monitoring based on local density estimation[J]. CIESC Journal, 2014, 65(8): 3071-3081.

参考文献 23

[1]	Tan Shuai, Wang Fuli, Peng Jun, Chang Yuqing, Wang S. Multimode process monitoring based on mode identification[J]. Ind. Eng. Chem. Res., 2012, 51: 374-388
[2]	Wold S, Esbensen K, Oeladi P. Principal component analysis[J]. Chemom. Intell. Lab. Syst., 1987, 2(1/2/3): 37-52
[3]	Nomikos P, Macgregor J F, Multi-way partial least squares in monitoring batch processes[J]. Chemom. Intell. Lab. Syst., 1995, 30(1): 97-108
[4]	Zhao Shijian, Zhang Jie, Xu Yongmao. Performance monitoring of process with multiple operation modes through multiple PLS models[J]. Journal of Process Control, 2006, 16(7): 763-772
[5]	Chu Young-Hwan, Qin S Joe, Han Chong-hun. Fault detection and operation mode identification based on pattern classification with variable selection[J]. Industrial and Engineering Chemistry Research, 2004, 43: 1701-1710
[6]	Ge Zhiqiang, Song Zhihuan. Robust monitoring and fault reconstruction based on variational inference component analysis[J]. Journal of Process Control, 2011, 21:462-474
[7]	Yu Jie. A new fault diagnosis method of multimode processes using Bayesian inference based Gaussian mixture contribution decomposition[J]. Engineering Applications of Artificial Intelligence, 2013, 26: 456-466
[8]	Yu Jie. A nonlinear kernel Gaussian mixture model based inferential monitoring approach for fault detection and diagnosis of chemical processes[J]. Chemical Engineering Science., 2012, 68: 506-519
[9]	Li Weihua, Yue H Henry, Sergio Valle-Cervantes, Qin S Joe. Recursive PCA for adaptive process monitoring[J]. Journal of Process Control, 2000, 10: 471-486
[10]	Teppola Pekka, Mujunen Satu-Pia, Minkkinen Pentti. Adaptive fuzzy C-means clustering in process monitoring[J]. Chemometrics and Intelligent Laboratory Systems, 1999, 45: 23-38
[11]	Jin Hyung Dae, Lee Young-Hak, Lee Gibaek, Han Chonghun. Robust recursive principal component analysis modeling for adaptive monitoring[J]. Industrial and Engineering Chemistry Research, 2006, 45: 696-703
[12]	Ge Zhiqiang, Song Zhihuan. Online monitoring of nonlinear multiple mode processes based on adaptive local model approach[J]. Control Engineering Practice, 2008, 16: 1427-1437
[13]	Xie Xiang, Shi Hongbo. Dynamic multimode process modeling and monitoring using adaptive Gaussian mixture models[J]. Industrial and Engineering Chemistry Research, 2012, 51: 5497-5505
[14]	Zhang Yingwei, Chai Tianyou, Li Zhiming, Yang Chunyu. Modeling and monitoring of dynamic processes[J]. IEEE Transactions on Neural Networks, 2012, 23: 277-284
[15]	He Q Peter, Jin Wang. Fault detection using the k-nearest neighbour rule for semiconductor manufacturing processes[J]. IEEE Transactions on Semiconductor Manufacturing, 2007, 20(4): 345-354
[16]	Ma Hehe, Hu Yi, Shi Hongbo. A novel local neighborhood standardization strategy and its application in fault detection of multimode processes[J]. Cemom. Intell. Lab. Syst., 2012, 118: 287-300
[17]	Zhang Ni(张妮), Tian Xuemin (田学民), Cai Lianfang(蔡连芳). Non-linear process fault detection method based on RISOMAP[J]. CIESC Journal (化工学报), 2013, 64(6): 2125-2130
[18]	He Xiaofei, Niyogi Partha. Locality preserving projections[J]. Advances in Neural Information Processing Systems, 2003, 16: 153-160
[19]	Latecki L J, Lazarevic A, Pokrajac D. Outlier Detection with Kernel Density Functions //Machine Learning and Data Mining in Pattern Recognition[M]. Springer Berlin Heidelberg, 2007: 61-75
[20]	Roussopoulos N, Kelley S, Vincent F. Nearest neighbor queries// ACM Sigmod Record[C]. New York,1995, 24(2): 71-79
[21]	Lee Jaeshin, Kang Bokyoung, Kang Suk-Ho. Integrating independent component analysis and local outlier factor for plant-wide process monitoring[J]. Journal of Process Control, 2011, 21: 1011-1021
[22]	Ge Zhiqiang, Song Zhihuan. Multimode process monitoring based on Bayesian method[J]. Chemometrics and Intelligent Laboratory Systems, 2009, 23: 636-650
[23]	Ricker N L. Decentralized control of the Tennessee Eastman challenge process[J]. Journal of Process Control, 1996, 6: 205-221

基于局部密度估计的多模态过程故障检测

Multimode process monitoring based on local density estimation

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献 23

相关文章 6

编辑推荐

Metrics

本文评价

[1]	许玉格, 邓文凯, 陈立定. 基于核函数的加权极限学习机污水处理在线故障诊断[J]. 化工学报, 2016, 67(9): 3817-3825.
[2]	韩敏, 张占奎. 基于改进核主成分分析的故障检测与诊断方法[J]. 化工学报, 2015, 66(6): 2139-2149.
[3]	吕小条, 宋冰, 谭帅, 侍洪波. 基于全变量信息的子空间监控方法[J]. 化工学报, 2015, 66(4): 1395-1401.
[4]	宋冰, 马玉鑫, 方永锋, 侍洪波. 基于LSNPE算法的化工过程故障检测[J]. 化工学报, 2014, 65(2): 620-627.
[5]	李健, 岳恒, 郭向红, 柴天佑. 铝酸钠叶滤过程控制仿真实验平台[J]. 化工学报, 2011, 62(8): 2089-2094.
[6]	徐芬;吴伟亮 . 喷雾场多参数层析算法的仿真效验与实验测量 [J]. CIESC Journal, 2006, 57(9): 2080-2085.