您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报(工学版) ›› 2017, Vol. 47 ›› Issue (5): 179-186.doi: 10.6040/j.issn.1672-3961.0.2017.181

• • 上一篇    下一篇

基于MLFDA的化工过程故障模式分类方法

王磊,邓晓刚*,曹玉苹,田学民   

  1. 中国石油大学(华东)信息与控制工程学院, 山东 青岛 266580
  • 收稿日期:2017-04-18 出版日期:2017-10-20 发布日期:2017-04-18
  • 通讯作者: 邓晓刚(1981— ),男,山东东营人,副教授,博士,主要研究方向为工业过程故障诊断技术. E-mail:dengxiaogang@upc.edu.cn E-mail:wangleipl@foxmail.com
  • 作者简介:王磊(1992— ),男,山东滨州人,硕士研究生,主要研究方向为多模态过程故障诊断. E-mail:wangleipl@foxmail.com
  • 基金资助:
    国家自然科学基金资助项目(61403418,61273160,21606256);山东省自然科学基金资助项目(ZR2014FL016,ZR2016FQ21,ZR2016BQ14);中央高校基本科研业务费专项资金资助项目(17CX02054);研究生创新工程资助项目(YCX2017058)

Multiblock local Fisher discriminant analysis for chemical process fault classification

WANG Lei, DENG Xiaogang*, CAO Yuping, TIAN Xuemin   

  1. College of Information and Control Engineering, China University of Petroleum, Qingdao 266580, Shandong, China
  • Received:2017-04-18 Online:2017-10-20 Published:2017-04-18

摘要: Fisher判别分析(FDA)是一种有效的化工过程故障模式分类方法,但是其忽视了数据局部结构信息的挖掘。针对该问题,提出一种多块局部Fisher判别分析(MLFDA)方法,以更有效地识别化工过程故障。从变量和样本两个维度来分析数据的局部结构特性。针对变量维度的局部信息挖掘问题,设计了一种基于变量与数据集主元空间的相关度的变量分块方法,将全局过程变量划分为多个局部变量块。进一步考虑到样本维度的局部结构特性,应用基于局部权重因子的局部Fisher判别分析(LFDA)为每个局部变量块构建分类器。提出一种基于分类性能加权的多分类器集成方法,以融合不同分类器的决策结果。在Tennessee Eastman过程上的仿真结果说明,MLFDA方法具有比传统的FDA和LFDA方法更低的故障误分类率。

关键词: Fisher判别分析, 故障模式分类, Tennessee Eastman过程, 局部Fisher判别分析, 多块局部Fisher判别分析

Abstract: Fisher discriminant analysis(FDA)was an effective chemical process fault classification method. However, the local data structure information was not investigated within traditional FDA method. To deal with this problem, a multiblock local Fisher discriminant analysis(MLFDA)method was proposed for more effective chemical process fault recognition. This method analyzed the local data structure characteristics from the variable-dimension and sample-dimension. To mine the local information in the variable-dimension, a variable block division method was designed based on the relevancy between the variables and the principal component subspace of the dataset, with which all the variables could be divided into several local variable blocks. Furthermore, considering the characteristics of local sample structure, the local FDA(LFDA)using local weighting factors was applied to construct classifier for each local variable block. An integrating strategy based on weighting classification performance weighting was presented to combine the results from different classifiers. Simulation results on Tennessee Eastman process showed that the proposed MLFDA method had a lower misclassification rate than traditional FDA and LFDA methods.

Key words: local Fisher discriminant analysis, multiblock local Fisher discriminant analysis, fault classification, Fisher discriminant analysis, Tennessee Eastman process

中图分类号: 

  • TP277
[1] QIN S J. Survey on data-driven industrial process monitoring and diagnosis[J]. Annual Reviews in Control, 2012, 36(2):220-234.
[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] YIN S, LI X, GAO H, et al. Data-based techniques focused on modern industry: an overview[J]. IEEE Transactions on Industrial Electronics, 2015, 62(1):657-667.
[4] 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 & Intelligent Laboratory Systems, 2013, 127(16):195-209.
[5] ZHANG Y, DU W, FAN Y, et al. Process fault detection using directional kernel partial least squares[J]. Industrial & Engineering Chemistry Research, 2015, 54(9):2509-2518.
[6] TIAN X, ZHANG X, DENG X, et al. Multiway kernel independent component analysis based on feature samples for batch process monitoring[J]. Neurocomputing, 2009, 72(7-9):1584-1596.
[7] PEP O, CAO Y. Nonlinear dynamic process monitoring using canonical variate analysis and kernel density estimations[J]. IEEE Transactions on Industrial informatics, 2010, 6(1):36-45.
[8] ZHAO S J, ZHANG J, XU Y. Performance monitoring of processes with multiple operating modes through multiple PLS models[J]. Journal of Process Control, 2006, 16(7):763-772.
[9] HUANG H, FENG H, PENG C. Complete local Fisher discriminant analysis with Laplacian score ranking for face recognition[J]. Neurocomputing, 2012, 89(10):64-77.
[10] ZHU Z, SONG Z. A novel fault diagnosis system using pattern classification on kernel FDA subspace[J]. Expert Systems with Application, 2011, 38(6):6895-6905.
[11] ZHONG S, WEN Q, GE Z. Semi-supervised Fisher discriminant analysis model for fault classification in industrial processes[J]. Chemometrics & Intelligent Laboratory Systems, 2014, 138:203-211.
[12] JIANG B, ZHU X, HUANG D, et al. A combined canonical variate analysis and Fisher discriminant analysis(CVA—FDA)approach for fault diagnosis[J]. Computers & Chemical Engineering, 2015, 77:1-9.
[13] SUGIYAMA M. Dimensionality reduction of multimodal labeled data by local fisher discriminant analysis[J]. Journal of Machine Learning Research, 2007, 8(1):1027-1061.
[14] YU J. Localized Fisher discriminant analysis based complex chemical process monitoring[J]. AIChE Journal, 2011, 57(7):1817-1828.
[15] REN S, SONG Z, YANG M, et al. A novel multimode process monitoring method integrating LCGMM with modified LFDA[J]. Chinese Journal of Chemical Engineering, 2015, 23(12):1970-1980.
[16] VAN M, KANG H. Wavelet kernel local fisher discriminant analysis with particle swarm optimization algorithm for bearing defect classification[J]. IEEE Transactions on Instrumentation & Measurement, 2015, 64(12):3588-3600.
[17] LI F, WANG J, CHYU M K, et al. Weak fault diagnosis of rotating machinery based on feature reduction with Supervised orthogonal local fisher discriminant analysis[J]. Neurocomputing, 2015, 168(C):505-519.
[18] CHIANG L H, RUSSELL E L, BRAATZ R D. Fault detection and diagnosis in industrial systems[M]. London: Springer, 2001.
[19] ADIL M, ABID M, KHAN A Q, et al. Exponential discriminant analysis for fault diagnosis[J]. Neurocomputing, 2016, 171(C):1344-1353.
[20] 张宇, 刘雨东, 计钊. 向量相似度测度方法[J]. 声学技术, 2009, 28(4): 532-536. ZHANG Yu, LIU Yudong, JI Zhao. Vector similarity measurement method[J]. Technical Acoustics, 2009, 28(4):532-536.
[21] DOWNS J J, VOGEL E F. A plant-wide industrial process control problem[J]. Computers & Chemical Engineering, 1993, 17(3):245-255.
[1] 程鑫,刘晗,王博,梁典,陈强. 基于双核处理器的主动磁悬浮轴承容错控制架构[J]. 山东大学学报(工学版), 2018, 48(2): 72-80.
[2] 周福娜,高育林,王佳瑜,文成林. 基于深度学习的缓变故障早期诊断及寿命预测[J]. 山东大学学报(工学版), 2017, 47(5): 30-37.
[3] 毛海杰,李炜,王可宏,冯小林. 基于自抗扰的多电机转速同步系统传感器故障切换容错策略[J]. 山东大学学报(工学版), 2017, 47(5): 64-70.
[4] 赵英弘,何潇,周东华. 一类含有传感器故障的网络化系统容错估计[J]. 山东大学学报(工学版), 2017, 47(5): 71-78.
[5] 秦利国,何潇,周东华. 一种时延多智能体系统的分布式编队[J]. 山东大学学报(工学版), 2017, 47(5): 79-88.
[6] 庞人铭,王波,叶昊,张海峰,李明亮. 基于PCA相似度和谱聚类相结合的高炉历史数据聚类[J]. 山东大学学报(工学版), 2017, 47(5): 143-149.
[7] 包塔拉,马剑,甘祖旺. 基于几何特征与流形距离的锂电池健康评估[J]. 山东大学学报(工学版), 2017, 47(5): 157-165.
[8] 叶晓丰, 王培良, 杨泽宇. 基于混合MPLS的多阶段过程质量预报方法[J]. 山东大学学报(工学版), 2017, 47(5): 246-253.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!