基于数据挖掘的宽厚板板凸度控制

来源期刊:中南大学学报(自然科学版)2019年第11期

论文作者:曹建国 江军 赵秋芳 何安瑞 李存福 孙旭东

文章页码:2743 - 2753

关键词:宽厚板轧机;数据挖掘;板形控制;凸度预测;随机森林;关联规则

Key words:wide and heavy plate mill; data mining; profile and flatness control; crown prediction; random forest; association rules

摘    要:针对某宽厚板CVC plus大型骨干工业轧机存在板凸度偏大且难以控制的问题,对现场轧制过程工艺及板凸度质量数据进行跟踪采集和分析,通过数据转换、提炼和集成建立标准的数据挖掘数据集,基于历史生产数据建立宽厚板板凸度随机森林预测模型,应用主成分分析以及减法聚类离散化方法进行数据的预处理,通过关联规则挖掘和控制变量影响力评估实现不同产品质量状态下关键控制变量的快速定位,并将其应用于典型规格宽厚板关键工艺参数调整策略;建立三维有限元耦合模型用于调整策略的仿真分析。研究结果表明:3种典型规格宽厚板板凸度分别下降30.9%,14.7%和23.9%;基于数据挖掘提出的调整策略可以有效改善板凸度控制情况,可为宽厚板板形质量控制研究提供参考。

Abstract: Considering that the crown of a wide and heavy plate CVC plus large mainstay industrial rolling mill was too large and difficult to control, rolling process and crown quality data were collected and analyzed. Standard data mining data sets were established through data conversion, extraction and integration. Based on historical production data, wide and heavy plate crown random forest prediction model was established. Principal component analysis and subtractive clustering discretization methods were used to preprocess the data. The rapid positioning of key control variables under different product quality conditions was realized by mining association rules and evaluating the influence of control variables. Then, it was applied to the adjustment strategy of key control process parameters for typical wide and heavy plates. A three-dimensional finite element coupling model was established for simulation and analysis of adjustment strategy. The results show that the crown of three typical wide and thick plates are reduced by 30.9%, 14.7% and 24.0%, respectively. The proposed adjustment strategy based on data mining can effectively improve the control of plate crown, and provides reference for the profile and flatness control of wide and heavy plate.

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