Journal of South China University of Technology (Natural Science Edition) ›› 2009, Vol. 37 ›› Issue (8): 1-7.

• Mechanical Engineering •     Next Articles

Rough Set-Based Models for On-Line Monitoring, Diagnosis and Control of Production Quality

Zhai Jing-mei  Xu Xiao  Yin Cun-fang  Xie Cun-xi   

  1. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China
  • Received:2008-07-03 Revised:2008-08-21 Online:2009-08-25 Published:2009-08-25
  • Contact: 翟敬梅（1967-），女，博士，副教授，主要从事生产系统建模和优化、数据挖掘、质量诊断和控制等的研究． E-mail:mejmzhai@seut．edu．cn
  • About author:翟敬梅（1967-），女，博士，副教授，主要从事生产系统建模和优化、数据挖掘、质量诊断和控制等的研究．
  • Supported by:

    广东省科技计划项目（20078010400049）

PDF

792

Abstract:

:Based on the statistical process control with two quality theories and the rough set theory, an integrated model of quality monitoring, diagnosis and control is proposed for the manufacturing process. In the investigation, the on-line quality monitoring and abnormal process diagnosis model based on the statistical process control with two quality theories is dealt with, and a new rough set-based quality diagnosis model is proposed to quantify the proba- bility and importance of the effects of manufacturing parameters on the quality, thus overcoming the information in- consistency and incompleteness inherent in manufacturing processes. Moreover, an optimal control algorithm of produefion quality oriented to user requirements is presented, which is then proved feasible. The correctness of the proposed models is finally verified by an application to yeast production.

Key words: quality monitoring, quality diagnosis, quality control, statistical process control, rough set theory