Rolling load model of CSP finishing mill group based on statistics method
YE Ben1,2,LI Li-xin2,HU Sheng-de 2
(1. Hot Rolling Company,WISDRI Engineering and Research Incorporation Limited,Wuhan 430223,Hubei,China 2. The State Key Laboratory of Refractories and Metallurgy,Wuhan University of Science and Technology, Wuhan 430081,Hubei,China)
Abstract:Data of CSP finishing mill group were analyzed by statistical method. The analysis of autocorrelation was carried out and the presence of serious autocorrelation phenomenon was revealed. The principal components analysis and ridge regression method were occupied to eliminate the autocorrelation. By fitting the Shida Shige model,the effects of autocorrelation were verified. Serious statistics method is used to eliminate some data. Modified Shida Shige model and multiple regression model were built. Then the precision of models is improved. The introduction of recrystallization gets a better accuracy. In the comparative test for 5 models,a set of 30 new data was used. Results show that the recrystallization model 5 has better accuracy.
收稿日期: 2015-01-15
出版日期: 2015-10-12
引用本文:
叶 奔,李立新,胡盛德. 统计学方法分析CSP精轧压力模型[J]. 钢铁, 2015, 50(10): 54-58.
YE Ben,,LI Li-xin,HU Sheng-de . Rolling load model of CSP finishing mill group based on statistics method. Iron and Steel, 2015, 50(10): 54-58.