论文部分内容阅读
为准确掌握超宽冷轧机不同宽度带钢的板形特征,以某2180 mm超宽冷轧机1900 mm宽度带钢实测板形数据为研究对象,借鉴“大数据”的思想,结合数据挖掘领域中聚类分析方法,提出基于网格和密度的板形特征聚类方法,并以此方法对几种典型带钢宽度的大量板形实测数据进行分析,得到不同宽度带钢的板形特征.以分段函数对板形特征进行多项式表达,得到不同宽度带钢的板形特征参数化分析结果.提出的基于网格和密度的板形特征聚类与分析方法,能够快速准确地对大量板形实测数据进行分析,提取出长期生产过程中板形缺陷特征并得到参数化表达,从而为冷连轧机,特别是超宽带钢冷连轧机的辊形改进和控制策略优化提供数据基础.
In order to accurately grasp the strip width characteristics of strip with different width of cold rolling mill, taking the measured data of 1900 mm width strip of a 2180 mm wide cold rolling mill as the research object and drawing on the idea of “big data” Data clustering method in the field of data mining, a mesh-based clustering method based on grid and density is proposed. Based on this method, a large number of strip-shaped measured data of several typical strip widths are analyzed, Shape feature.The piece shape feature is expressed by polynomial with piecewise function to get the parametric analysis results of the plate shape characteristics of strip with different widths.The proposed clustering and analysis method of plate shape feature based on grid and density can quickly and accurately A large number of plate-shaped measured data were analyzed to extract the characteristics of plate-shaped defects in the long-term production process and get parametric expression, so as to provide the data foundation for the roll shape improvement and control strategy optimization of the tandem cold mill .