不能及时获得收得率影响因素的检测数据是钢包精炼炉元素收得率预报的难点之一。为了解决该问题,首先通过机制分析,在可测变量中选取并创建可以间接表达收得率影响因素的变量,然后将这些变量作为模型的输入,使用支持向量机方法建立元素收得率预报模型。在试验中,将本方法与已有方法进行了比较,比较结果表明本方法所建模型有较高的预报精确度与命中率,更适合于在生产中使用。
The influencing factors of element recovery could not be obtained instantly,which was one of the difficulties of recovery prediction in ladle furnace（LF）.In order to solve this problem,through mechanism analysis,some measurable variables were selected and some new variables that can express the factors of element recovery indirectly were created by using measurable variables.Then,the selected and created variables were used as inputs of element recovery prediction model that was established using support vector regression（SVR）.In the experiment,the proposed method was compared with existing recovery prediction methods.The results show that the proposed method has higher accuracy and hit rate,and is more suitable in production.
参考文献
| [1] | 于鹏,战东平,姜周华,李大量,尹小东,马志刚.LF精炼终点成分预报模型开发[J].材料与冶金学报,2006(01):20-22. |
| [2] | 高宪文,张傲岸,魏庆来.基于神经网络的钢包精炼终点预报[J].东北大学学报（自然科学版）,2005(08):726-728. |
| [3] | 李洪桂.冶金原理[M].北京:科学出版社,2004 |
| [4] | 李晶.LF精炼技术[M].北京:冶金工业出版社,2009 |
| [5] | Vapnik V N.The Nature of Statistical Learning Theory[M].Beilin:Springer-Verlag,1995 |
| [6] | Vapnik V N.Statistical Learning Theory[M].New York:John Wiley and Sons,Inc,1998 |
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%