基于时域分析的拉坯阻力曲线异常特征提取及在漏钢预报上的应用  

Abnormal characteristics capture in the mould friction curve and its application in the breakout prediction of continuous casting based on time-region analysis

作　　者：赵辉[1] 武冬蕾[2] 刘鲁源[1] 王红君[2] 岳有军[2] 

机构地区：[1]天津大学,天津300072 [2]天津理工大学,天津300191

出　　处：《制造业自动化》2005年第5期61-64,共4页Manufacturing Automation

基　　金：天津市自然科学基金重点资助项目(003800811);天津市教委科技发展基金资助项目(20041705)

摘　　要：介绍了板坯连铸结晶器拉坯阻力异常特征提取及在连铸漏钢预报中应用的研究。在对漏钢事故前拉坯阻力时域异常特征分析总结的基础上,针对两种异常特征模式——尖脉冲和斜坡设计了两种特征提取算法——幅值变化判断法和改进的过零率判断法,实际应用证明了算法的有效性。This paper presents the two capturing algorithms of abnormal characteristics in the friction curve for the slab continuous caster, with the discussion of their application in the breakout prediction of continuous casting. Based on the analysis and summary of the abnormal time- region characteristics in the friction curve before breakout, ample-change judgement and modified pass-zero-ratio judgement algorithms are designed to capture the two abnormal modes, sharp pulse and big ramp. These algorithms are proved valid by the practical application.

关 键 词：连铸结晶器 拉坯阻力 在线监测 漏钢预报 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]