• DocumentCode
    1895718
  • Title

    Process Monitoring for Integration of SPC and APC Based on BP Neural Network

  • Author

    Yu, Jianli ; Zhang, Zongwei ; Xu, Liang

  • Author_Institution
    Sch. of Electr. & Inf. Eng., Zhongyuan Univ. of Technol., Zhengzhou, China
  • Volume
    1
  • fYear
    2009
  • fDate
    10-11 Oct. 2009
  • Firstpage
    378
  • Lastpage
    381
  • Abstract
    Aimed at an integration system of SPC (statistical process control) and APC (automatic process control), the effect of APC on SPC detection capability is analyzed. In this paper, the process disturbance is assumed to be an ARMA (1, 1) process. BP Neural Network with good capability of mode identification is used to substitute traditional SPC technology. Then an integrated design methodology has been developed for APC and BP neural network monitoring for the purpose of process special cause. A simulation result reveals that BP Neural Networks can detect special cause quickly and ameliorate SPC monitoring capability effectively. The ARL performance is studied. Lots of simulation experiments reveal that the proposed design approaches outperform the traditional integrated scheme of SPC and APC.
  • Keywords
    autoregressive moving average processes; backpropagation; neurocontrollers; process monitoring; statistical process control; BP neural network; automatic process control; detection capability; integrated design methodology; mode identification; process disturbance; process monitoring; statistical process control; Analytical models; Automation; Computer networks; Control systems; Intelligent networks; Monitoring; Neural networks; Process control; Production; Quality control; BP Neural Network; Integration system of SPC and APC; control chart; special cause;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Computation Technology and Automation, 2009. ICICTA '09. Second International Conference on
  • Conference_Location
    Changsha, Hunan
  • Print_ISBN
    978-0-7695-3804-4
  • Type

    conf

  • DOI
    10.1109/ICICTA.2009.99
  • Filename
    5287633