| 基于主元分析的转炉终点ESN静态预测模型 |
| |
| 引用本文： | 王玉昆,张勇.基于主元分析的转炉终点ESN静态预测模型[J].鞍山科技大学学报,2009(4):360-364. |
| |
| 作者姓名： | 王玉昆 张勇 |
| |
| 作者单位： | 辽宁科技大学电子与信息工程学院; |
| |
| 摘 要： | 以主元分析方法和新型ESN(回声状态网络)算法为核心,研究了转炉终点静态预测模型。通过对某钢厂转炉生产数据的主元分析,建立了ESN模型,同时将ESN模型与传统的BP和RBF神经网络模型进行了对比研究。结果表明,使用ESN建立的模型比传统的BP网络模型和RBF网络模型,在钢水温度预测方面精度分别提高了0.85%和0.45%,在钢水碳质量分数预测方面精度分别提高了0.45%和0.19%,能够有效的对转炉终点碳含量和温度进行预测,从而为转炉炼钢过程提供更准确的操作指导。
|
| 关 键 词： | 转炉炼钢 命中率 主元分析 回声状态网络 |
ESN static prediction model for BOF end-point based on PCA |
| |
| WANG Yu-kun,ZHANG Yong.ESN static prediction model for BOF end-point based on PCA[J].Journal of Anshan University of Science and Technology,2009(4):360-364. |
| |
| Authors: | WANG Yu-kun ZHANG Yong |
| |
| Institution: | School of Electronic and Information Engineering;University of Science and Technology Liaoning;Anshan 114051;China |
| |
| Abstract: | This paper is concerned with the BOF end-point static prediction model based on PCA algorithm and new ESN(echo state network).The PCA processed is utlized to established the ESN model by anlasying BOF production data of an steelwork.In addition,a comparative study was done with the traditional BP and RBF neural network model.The results show that the ESN model achieves encouraging 0.85% and 0.45% in the aspect of the temperature prediction of molten steel,and 0.45% and 0.19% in the aspect of the carbon mass... |
| |
| Keywords: | BOF hit rate PCA ESN neural network |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
