期刊文献+

基于自适应主成分分析的化工过程在线监测 被引量:2

Online Monitoring of Chemical Process Based on Adaptive Principal Component Analysis
下载PDF
导出
摘要 主成分分析(principal component analysis,PCA)应用于过程监测时,不适当的成分选择方法会导致变异特征被分散或被淹没从而影响监测性能。针对这个问题提出了成分的自适应选择方法并用于过程监测,即自适应主成分分析(adaptive principal component analysis,APCA)。自适应主成分应用于过程监测时主要包括3个步骤:首先,在离线建模时基于载荷矩阵通过欧氏距离计算各个成分的相似性,并基于每个成分选出与其相似性较高的成分构成多个成分子空间;其次,在线监测时基于在线样本的各成分通过核密度估计计算各个成分的变异概率,选择出变异概率最高的成分作为特征成分;最后,挑选出与特征成分对应的成分子空间,并构造T 2统计量。通过数值仿真案例和田纳西伊斯曼(tennessee eastman,TE)过程证明了提出方法APCA的有效性。 When Principal Component Analysis(PCA)was applied to process monitoring,improper component selection method would cause variation characteristics to be dispersed or submerged,thus affecting monitoring performance.In order to solve this problem,An adaptive selection method of components called Adaptive Principal Component Analysis(APCA)was proposed and applied it to process monitoring.The application of adaptive principal components to process monitoring mainly included three steps.Firstly,the similarity of each component was caculated based on the load matrix through Euclidean distance during offline modeling,and components with high similarity to each component was selected to form multiple molecular spaces.Secondly,during on-line monitoring,the variation probability of each component was calculated by kernel density estimation based on each component of the on-line sample,and the component with the highest variation probability was selected as the characteristic component.Finally,the molecular space corresponding to CC was selected and statistics were constructed.The result of numerical simulation and Tennessee Eastman(TE)process proved the effectiveness of the proposed APCA.
作者 吕照民 周革 苗晨 LYU Zhaomin;ZHOU Ge;MIAO Chen(School of Urban Rail Transportation,Shanghai University of Engineering Science,Shanghai 201620,China;Shanghai Electro-mechanical Engineering Institute,Shanghai 201109,China;School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处 《郑州大学学报(工学版)》 CAS 北大核心 2020年第1期44-48,共5页 Journal of Zhengzhou University(Engineering Science)
基金 国家自然科学基金青年基金资助项目(61703275) 上海市青年科技英才杨帆计划(18YF1409200) 上海工程技术大学人才计划项目——展翅计划。
关键词 过程监测 主成分分析 子空间 自适应 process monitoring principal component analysis subspace adaptive
  • 引文网络
  • 相关文献

参考文献4

二级参考文献19

  • 1赵忠盖,刘飞.因子分析及其在过程监控中的应用[J].化工学报,2007,58(4):970-974. 被引量:24
  • 2Chiang L H, Russell E L, Braatz R D. Fault Detection and Diagnosis in Industrial Systems. London: Springer- Verlag, 2001
  • 3Kim D, Lee I B. Process monitoring based on probabilistic PCA. Chemometrics and Intelligent Laboratory Systems, 2003, 67:109-123
  • 4Hyvarinen A, Oja E. Independent component analysis: algorithms and applications. Neural Network, 2000, 13: 411-430
  • 5Kano M, Tanaka S, Hasebe S, Hashimoto I, Ohno H. Monitoring independent components for fault detection. AIChEJournal, 2003, 49 (4): 969-976
  • 6Xie L, Wu J. Gobal optimal ICA and its application in MEG data analysis. Neurocomputing, 2006, 69:2438-2442
  • 7He Ning(何宁).Research on performance monitoring and fault diagnosis for process industry based on ICA PCA method[D]. Hangzhou: Zhejiang University, 2004: 31-54
  • 8Hyvarinen A. A fast fixed point algorithm for independent component analysis. Neural Computation, 1997, 9:1483-1492
  • 9Chen Q, Kruger U, Andrew T Y Leung. Regularised kernel density estimation for clustered process data. Control Engineering Practice, 2004, 12:267-274
  • 10Zhong Mingjun(钟明军).Some algorithms for independent component analysis and their application to fMRI data analysis [D]. Dalian: Dalian University of Technology, 2004: 31-42

共引文献48

同被引文献9

引证文献2

;
使用帮助 返回顶部