Design of regression model-based automatic process control with reduced adjustment frequency

Authors

  • Liang Ye,

    1. Department of Industrial Engineering and Logistic Engineering, School of Mechanical Engineering, Shanghai Jiao-tong University, Shanghai 200240, People's Republic of China
    Search for more papers by this author
  • Ershun Pan,

    1. Department of Industrial Engineering and Logistic Engineering, School of Mechanical Engineering, Shanghai Jiao-tong University, Shanghai 200240, People's Republic of China
    Search for more papers by this author
  • Jianjun Shi

    Corresponding author
    1. School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0205, U.S.A.
    • School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0205, U.S.A.
    Search for more papers by this author

Abstract

Automatic process control based on a regression model has been adopted as one of the important techniques to improve product quality in manufacturing processes. Though more frequent adjustments generally produce a superior control performance, it may also increase control cost and impair control applicability. In this paper, the concepts of quality margin and self-compensation of noise change are introduced. Based on these concepts, a control strategy is proposed which is capable of ensuring an acceptable process performance with a reduced adjustment frequency. A case study of leaf spring forming process is conducted to compare the control performance and control adjustment frequency between the proposed approach and the existing methods. Some properties of the proposed control law are also studied. The proposed method is implemented in a hot steel rolling process to demonstrate the applicability of the proposed method. Copyright © 2009 John Wiley & Sons, Ltd.

Ancillary