冷轧平整机工作辊表面粗糙度衰减模型

发布时间：2018-05-09 14:25

本文选题：表面粗糙度 + 灰色关联度分析　；参考：《钢铁》2015年06期

【摘要】：冷轧平整机的工作辊直接和带钢接触,其表面粗糙度衰减情况对带钢成品的板形和表面质量有重大影响。因此,分析轧辊磨损机制,对轧辊表面粗糙度的衰减进行精确预测十分必要。首先采用灰色关联度分析对影响平整机工作辊表面粗糙度磨损的因素进行分析,确定了工作辊表面粗糙度评估指标体系。进而应用优化在线稀疏最小二乘支持向量回归模型对冷轧平整机的上工作辊表面粗糙度进行在线预测。通过预测误差准则实现系统的前向递推,采用FLOO(fast leave one out)的修剪算法实现其后向删减,并且采用最速下降法实现了2个超参数的在线优化。经过仿真研究表明,系统预测的绝对误差平均值为0.014 9,与其他方法相比具有明显的优越性,并且系统具有在线自适应的能力,能够随着时间而进化。
[Abstract]:The work roll of the cold rolling mill is in direct contact with the strip, and the attenuation of the surface roughness has a great influence on the shape and surface quality of the finished strip. Therefore, it is necessary to accurately predict the surface roughness attenuation by analyzing the roll wear mechanism. Firstly, the grey correlation analysis was used to analyze the factors affecting the surface roughness of the work roll, and the evaluation index system of the surface roughness of the work roll was determined. Furthermore, the surface roughness of the upper work roll of the cold rolling mill is predicted by using the optimal online sparse least square support vector regression model. The prediction error criterion is used to realize the forward recursion of the system, the pruning algorithm of FLOO(fast leave one out) is used to realize the backward pruning, and the on-line optimization of two superparameters is realized by the steepest descent method. The simulation results show that the average absolute error predicted by the system is 0.014, which is superior to other methods, and the system has the ability of online adaptation and can evolve over time.
【作者单位】：燕山大学工业计算机控制工程河北省重点实验室;燕山大学国家冷轧板带装备及工艺工程技术研究中心;新兴铸管股份有限公司格板部;
【基金】：国家自然科学基金钢铁联合基金资助项目(U1260203) 河北省高等学校创新团队领军人才培育计划资助项目(LJRC013)
【分类号】：TG333

【参考文献】