影响高炉铁水硅含量的因素往往复杂多变,影响程度不一.采用鱼骨分析法收集所有可能对硅含量产生影响的因素,经过相关分析和特征选择,最终选取6个参数作为模型的输入参数.采用改进的粒子群优化算法对支持向量机(SVM)中的参数进行优化,提出基于变邻域粒子群(VNPSO)优化 SVM的铁水硅含量预测模型.通过钢厂的实际生产数据进行验证,平均相对误差达到0.69％,平均绝对误差达到3.4×10－3,模型具有很高的预测精度.同时,绘制铁水中硅含量控制图,分析硅含量波动情况,并依此模型给出硅含量稳定性控制措施.
There are some complicated and changeable factors influencing silicon content in the furnace.All pos-sible factors influencing silicon content were collected by fishbone analyzing and finally six parameters were se-lected as the input parameters of the model by correlation analysis and feature selection.An improved particle swarm optimization (VNPSO)algorithm was used to optimize the parameters in support vector machine (SVM)and a prediction model of silicon content in hot metal based on variable neighborhood particle swarm opti-mization was proposed.The average relative error is 0.69% and the average absolute error is 3.4×10-3 by com-paring calculating results with the actual production data of steel.At the same time,fluctuation trend of silicon content is analyzed by the control chart of silicon content,and the measure of controlling the stability of silicon content is given at last.
参考文献
| [1] | Vapnik Vladimir N..Overview of statistical learning theory[J].IEEE Transactions on Neural Networks,19995(5):988-999. |
| [2] | 王义康;刘祥官.基于FCM的多支持向量机模型在高炉炉温预测中的应用[J].冶金自动化,2012(3):18-23. |
| [3] | 崔桂梅;孙彤;张勇.支持向量机在高炉铁水温度预测中的应用[J].控制工程,2013(5):809-812,817. |
| [4] | 崔桂梅;李静;张勇;卢俊慧;马祥.高炉铁水温度的多元时间序列建模和预测[J].钢铁研究学报,2014(4):33-37. |
| [5] | 谈霖;张宗旺;任冰朗;张海滨.基于支持向量机的铁水硅含量的预测[J].钢铁研究,2016(1):6-9. |
| [6] | 朱海梅;吴永萍.一种高速收敛粒子群优化算法[J].控制与决策,2010(1):20-24,30. |
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%