CIESC Journal

• SYSTEM ENGINEERING • 上一篇    下一篇

基于非负频谱分解的厂级多重振荡源的分离研究

夏春明a; 郑建荣a; J.Howellb   

  1. a Centre for Mechatronics Engineering, East China University of Science & Technology,
    Shanghai 200237, China
    b Department of Mechanical Engineering, University of Glasgow, Glasgow G12 8QQ, UK
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-06-28 发布日期:2007-06-28
  • 通讯作者: 夏春明

Isolation of whole-plant multiple oscillations via non-negative spectral decomposition

XIA Chunminga; ZHENG Jianronga; J.Howellb   

  1. a Centre for Mechatronics Engineering, East China University of Science & Technology,
    Shanghai 200237, China
    b Department of Mechanical Engineering, University of Glasgow, Glasgow G12 8QQ, UK
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-06-28 Published:2007-06-28
  • Contact: XIA Chunming

摘要: Constrained spectral non-negative matrix factorization (NMF) analysis of perturbed
oscillatory process control loop variable data is performed for the isolation of multiple
plant-wide oscillatory sources. The technique is described and demonstrated by analyzing
data from both simulated and real plant data of a chemical process plant. Results show that
the proposed approach can map multiple oscillatory sources onto the most appropriate
control loops, and has superior performance in terms of reconstruction accuracy and
intuitive understanding compared with spectral independent component analysis (ICA).

关键词: process monitoring;multiple oscillations;non-negative matrix factorization;sparse; spectral analysis;fault isolation

Abstract: Constrained spectral non-negative matrix factorization (NMF) analysis of perturbed
oscillatory process control loop variable data is performed for the isolation of multiple
plant-wide oscillatory sources. The technique is described and demonstrated by analyzing
data from both simulated and real plant data of a chemical process plant. Results show that
the proposed approach can map multiple oscillatory sources onto the most appropriate
control loops, and has superior performance in terms of reconstruction accuracy and
intuitive understanding compared with spectral independent component analysis (ICA).

Key words: process monitoring, multiple oscillations, non-negative matrix factorization, sparse, spectral analysis, fault isolation