东北大学学报:自然科学版 ›› 2017, Vol. 38 ›› Issue (3): 366-369.DOI: 10.12068/j.issn.1005-3026.2017.03.013

• 材料与冶金 • 上一篇    下一篇

基于软测量模型的粗轧厚度预测方法

彭文, 马更生, 龚殿尧, 张殿华   

  1. (东北大学 轧制技术及连轧自动化国家重点实验室, 辽宁 沈阳110819)
  • 收稿日期:2015-10-14 修回日期:2015-10-14 出版日期:2017-03-15 发布日期:2017-03-24
  • 通讯作者: 彭文
  • 作者简介:彭文(1987-),男,山东青州人,东北大学博士后研究人员; 张殿华(1963-),男,内蒙古赤峰人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(51504061).

Prediction Method of Rough Rolling Thickness Based on the Soft Sensor Model

PENG Wen, MA Geng-sheng, GONG Dian-yao, ZHANG Dian-hua   

  1. State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110819, China.
  • Received:2015-10-14 Revised:2015-10-14 Online:2017-03-15 Published:2017-03-24
  • Contact: PENG Wen
  • About author:-
  • Supported by:
    -

摘要: 为提高热连轧粗轧过程各道次厚度控制精度, 满足道次动态修正的要求,提出了一种简单有效的厚度软测量模型. 基于典型粗轧区的仪表配置,模型从轧件跟踪、数据处理等方面进行了针对性的研究,提出了偶数道次立辊辊缝渐变的策略用于可逆轧制过程中的宽度测量,并通过模型预测得到了轧制过程中各道次的出口厚度. 现场实际应用表明,通过软测量模型预报的各道次厚度与实际测量结果吻合较好,各道次的厚度预报偏差在±0.10mm范围内达到95.2%,满足了现场生产需求.

关键词: 热连轧, 粗轧, 软测量模型, 厚度预测, 控制精度

Abstract: In order to improve the thickness control precision and meet the demand of the pass dynamic correction in rough rolling process, a simple and effective thickness soft sensor model was proposed. The study of rolling tracking and data processing was carried out based on the instrument arrangement in typical rough rolling. The strategy of gradual vertical roll-gap was proposed to measure the width in the even pass. And the thickness of each pass was obtained by the soft sensor model. The application results showed that the thickness is in a good agreement with the actual measured ones. The predicted precision error in the range of ±0.10mm reaches 95.2%, which can meet the demand of the production.

Key words: hot rolling, rough rolling, soft sensor model, thickness prediction, control precision

中图分类号: