基于过程参数控制的烧结矿质量预测模型
|
摘要
针对烧结过程非线性、强耦合性和大时滞的特点,从过程参数控制的角度对烧结工艺进行了总体分析,确定了烧结矿性能评价指标及其主要影响参数,在此基础上提出了一种带动量项和变学习率的BP神经网络算法,建立了烧结矿质量预测模型。仿真实验结果表明,模型具有较强的自学习功能和较高的预测精度,用拓扑结构为15-25-4的BP神经网络和0.65×10-3的网络误差进行训练,模型的预报命中率在81.25%以上,充分验证了基于过程参数控制的烧结矿质量预测模型的准确性和有效性。
Aiming at characteristics of non-linearity,strong coupling and long time-delay in the sintering process,an overall analysis was conducted for the sintering process from the perspective of process parameter control. Thus,the sintered ore properties evaluation index and the main impacting parameters for it were all obtained. Then,a BP neural network algorithm with momentum and variable learning rate was proposed,with which a quality prediction model for the sintered ore established. The following simulation experimental results showed that the model has a higher prediction precision and stronger self-learning ability. The predictive hit-ratio of random samples was over 81.25% by adopting BP neural network with the structure of 15-25-4 and network error of 0. 65 × 10-3,which verified the accuracy and effectiveness of this quality prediction model on the basis of process parameter control.
Aiming at characteristics of non-linearity,strong coupling and long time-delay in the sintering process,an overall analysis was conducted for the sintering process from the perspective of process parameter control. Thus,the sintered ore properties evaluation index and the main impacting parameters for it were all obtained. Then,a BP neural network algorithm with momentum and variable learning rate was proposed,with which a quality prediction model for the sintered ore established. The following simulation experimental results showed that the model has a higher prediction precision and stronger self-learning ability. The predictive hit-ratio of random samples was over 81.25% by adopting BP neural network with the structure of 15-25-4 and network error of 0. 65 × 10-3,which verified the accuracy and effectiveness of this quality prediction model on the basis of process parameter control.
引文
[1]范晓慧,冯婧,陈许玲,等.烧结矿化学成分预报模型与控制指导专家系统[J].矿冶工程,2011,31(4):77-85.
[2]张华,黄昌先,赵刚,等.基于辅料运行特性的烧结矿质量预测模型[J].钢铁研究学报,2012,24(4):14-18.
[3] Fan X H,Li Y,Chen X L. Prediction of iron ore sintering characters on the basis of regression analysis and artificial neural network[J].Energy Procedia,2012,16(B):769-776.
[4] Wu M,Chen X X,Cao W H,et al. An intelligent integrated optimization system for the proportioning of iron ore in a sintering process[J]. Journal of Process Control,2014,24(1):182-202.
[5]向婕.铁矿石烧结过程智能集成优化控制技术及其应用研究[D].长沙:中南大学信息科学与工程学院,2010.
[6] Fan X H,Huang X X,Chen X L,et al. Research and development of the intelligent control of iron ore sintering process based on fan frequency conversion[J]. Ironmaking&Steelmaking,2016,43(7):488-493.
[7]吕庆,黄宏虎,曹丽华,等.原矿中Fe O含量对烧结矿的影响[J].烧结球团,2014,39(5):1-4.
[8]王筱留.钢铁冶金学(3版)[M].北京:冶金工业出版社,2013.
[9]陈革,王瑞军,沈茂森.包钢自熔性复合烧结矿试验研究[J].矿冶工程,2013,33(3):98-103.
[10]赵凯,李杰,张玉柱,等.邢钢烧结原料基础特性及优化配矿试验研究[J].矿冶工程,2016,36(4):88-92.
[11]杨景明,顾佳琪,闫晓莹,等.基于改进遗传算法优化BP网络的轧制力预测研究[J].矿冶工程,2015,35(1):111-115.
[12]田宝亮,牛培峰.基于GA-BP神经网络的弯辊力预设定模型研究与应用[J].矿冶工程,2018,38(1):111-114.
[13] Li Z P,Fan X H,Guo C,et al. Optimization of iron ore sintering process based on ELM model and multi-criteria evaluation[J]. Neural Computing and Applications,2017,28(8):2247-2253.
[2]张华,黄昌先,赵刚,等.基于辅料运行特性的烧结矿质量预测模型[J].钢铁研究学报,2012,24(4):14-18.
[3] Fan X H,Li Y,Chen X L. Prediction of iron ore sintering characters on the basis of regression analysis and artificial neural network[J].Energy Procedia,2012,16(B):769-776.
[4] Wu M,Chen X X,Cao W H,et al. An intelligent integrated optimization system for the proportioning of iron ore in a sintering process[J]. Journal of Process Control,2014,24(1):182-202.
[5]向婕.铁矿石烧结过程智能集成优化控制技术及其应用研究[D].长沙:中南大学信息科学与工程学院,2010.
[6] Fan X H,Huang X X,Chen X L,et al. Research and development of the intelligent control of iron ore sintering process based on fan frequency conversion[J]. Ironmaking&Steelmaking,2016,43(7):488-493.
[7]吕庆,黄宏虎,曹丽华,等.原矿中Fe O含量对烧结矿的影响[J].烧结球团,2014,39(5):1-4.
[8]王筱留.钢铁冶金学(3版)[M].北京:冶金工业出版社,2013.
[9]陈革,王瑞军,沈茂森.包钢自熔性复合烧结矿试验研究[J].矿冶工程,2013,33(3):98-103.
[10]赵凯,李杰,张玉柱,等.邢钢烧结原料基础特性及优化配矿试验研究[J].矿冶工程,2016,36(4):88-92.
[11]杨景明,顾佳琪,闫晓莹,等.基于改进遗传算法优化BP网络的轧制力预测研究[J].矿冶工程,2015,35(1):111-115.
[12]田宝亮,牛培峰.基于GA-BP神经网络的弯辊力预设定模型研究与应用[J].矿冶工程,2018,38(1):111-114.
[13] Li Z P,Fan X H,Guo C,et al. Optimization of iron ore sintering process based on ELM model and multi-criteria evaluation[J]. Neural Computing and Applications,2017,28(8):2247-2253.