基于批次回归系数的热轧带钢头部拉窄过程监控与诊断

来源期刊:中南大学学报(自然科学版)2018年第3期

论文作者:何飞 孙勇 杨德斌

文章页码:574 - 583

关键词:批次数据分析;偏最小二乘回归;回归系数;过程监控;质量诊断

Key words:batch data analysis; partial least squares regression; regression coefficients; process monitoring; quality diagnosis

摘    要:针对目前热轧带钢生产过程数据分析主要利用工艺参数和产品质量的平均值,忽略带钢长度方向上的变异信息,提出一种新的三维生产过程数据的监控框架。首先建立热连轧机组精轧过程中每块带钢的工艺变量与宽度间的偏最小二乘模型,获得所有批次的回归系数并组成二维数据矩阵,利用回归系数矩阵建立主成分分析模型进行监控和诊断。研究结果表明:该方法可以有效获取过程变量对质量的影响关系、实现过程监控,并有效给出质量异常的原因。

Abstract: Considering that the current data analysis methods usually use the mean value of process and quality parameters in hot rolled strip production without including the variation information in the longitudinal direction of the strip, a novel framework was introduced for process monitoring using three-way dataset. Firstly, the partial least squares model between process variables and width of each strip after finishing hot rolling was built. And then, a two-way matrix was obtained, which consisted of regression coefficients of all batches. Finally, regression coefficients matrix was used to process monitoring and diagnosis based on principal component analysis. The results show that the new method can not only effectively obtain the relationship between the process and quality parameter, but also finish process monitoring and explain why there appears abnormal quality.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号