东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2008, Vol. 29 ›› Issue (12): 1718-1720.DOI: -

• 论著 • 上一篇    下一篇

自学习在模拟连续退火过程控制中的应用

花福安;王文乐;李建平;刘相华;   

  1. 东北大学轧制技术及连轧自动化国家重点实验室;
  • 收稿日期:2013-06-22 修回日期:2013-06-22 出版日期:2008-12-15 发布日期:2013-06-22
  • 通讯作者: Hua, F.-A.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(50527402)

Application of self-learning to simulation for process control in continuous annealing

Hua, Fu-An (1); Wang, Wen-Le (1); Li, Jian-Ping (1); Liu, Xiang-Hua (1)   

  1. (1) The State Key Laboratory of Rolling Technology and Automation, Northeastern University, Shenyang 110004, China
  • Received:2013-06-22 Revised:2013-06-22 Online:2008-12-15 Published:2013-06-22
  • Contact: Hua, F.-A.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

701

摘要： 以理论分析为基础,构建了模拟连续退火的冷却过程控制数学模型.为了提高自学习效率,提出了将冷却过程中辐射散热等效转化为对流散热系数的观点.模型对试样温度进行层别的划分,并且针对各个温度层别,采用指数平滑处理的方法确定各个温度层别的调整系数.在实际冷却过程中,过程控制系统根据自学习得到的各个温度层别的调整系数实时地调整冷却气体的流量.实际应用表明,以自学习为基础的过程控制系统可使模拟连续退火的冷却过程的温度精度控制在10℃以内.

关键词: 连续退火, 模拟, 过程控制, 自学习, 冷却

Abstract: Based on theoretical analysis, a mathematical model was developed to simulate the cooling process control in continuous annealing. To improve the efficiency of self-learning, a viewpoint was suggested that the radiant heat dissipation can be transformed equivalently into the adjustable convective heat dissipation coefficients in cooling process. In the model, the sample temperature is layered and the exponential smoothing is introduced to determine the adjusting coefficients for each and all temperature layers. In the actual cooling process, the flow rate of cooling gas can be adjusted by virtue of the process control system according to the adjusting coefficients for different layers. Application results indicated that process control system based on self-learning enable the temperature accuracy to be controlled within 10°C in the simulation of cooling process for continuous annealing.

中图分类号: