SCI和EI收录∣中国化工学会会刊

Chinese Journal of Chemical Engineering ›› 2012, Vol. 20 ›› Issue (6): 1121-1127.

• • 上一篇    下一篇

Phase Analysis and Identification Method for Multiphase Batch Processes with Partitioning Multi-way Principal Component Analysis (MPCA) Model*

董伟威1, 姚远2, 高福荣1   

  1. 1. National Engineering Research Center for Industrial Automation(South China), Guangzhou 511458, China;
    2. Department of Chemical Engineering, National Tsing Hua University, Hsinchu 30013, China
  • 收稿日期:2012-05-27 修回日期:2012-08-27 出版日期:2012-12-28 发布日期:2012-12-28
  • 通讯作者: GAO Furong,E-mail:kefgao@ust.hk
  • 基金资助:
    Supported by the Guangzhou Scientific and Technological Project (2012J5100032);Nansha District Independent Innovation Project (201103003)

Phase Analysis and Identification Method for Multiphase Batch Processes with Partitioning Multi-way Principal Component Analysis (MPCA) Model*

DONG Weiwei1, YAO Yuan2, GAO Furong1   

  1. 1. National Engineering Research Center for Industrial Automation(South China), Guangzhou 511458, China;
    2. Department of Chemical Engineering, National Tsing Hua University, Hsinchu 30013, China
  • Received:2012-05-27 Revised:2012-08-27 Online:2012-12-28 Published:2012-12-28
  • Supported by:
    Supported by the Guangzhou Scientific and Technological Project (2012J5100032);Nansha District Independent Innovation Project (201103003)

摘要: Multi-way principal component analysis (MPCA) is the most widely utilized multivariate statistical process control method for batch processes.Previous research on MPCA has commonly agreed that it is not a suitable method for multiphase batch process analysis.In this paper,abundant phase information is revealed by way of partitioning MPCA model,and a new phase identification method based on global dynamic information is proposed.The application to injection molding shows that it is a feasible and effective method for multiphase batch process knowledge understanding,phase division and process monitoring.

关键词: batch process, multi-way principal component analysis, multiphase, process monitoring

Abstract: Multi-way principal component analysis (MPCA) is the most widely utilized multivariate statistical process control method for batch processes.Previous research on MPCA has commonly agreed that it is not a suitable method for multiphase batch process analysis.In this paper,abundant phase information is revealed by way of partitioning MPCA model,and a new phase identification method based on global dynamic information is proposed.The application to injection molding shows that it is a feasible and effective method for multiphase batch process knowledge understanding,phase division and process monitoring.

Key words: batch process, multi-way principal component analysis, multiphase, process monitoring