基于投影寻踪回归的铜熔炼过程中关键工艺指标预测模型研究

详细信息    查看全文 | 推荐本文 |

英文篇名：Projection Pursuit Regression-Based Key Process Indictors Predicting Model for Copper Flash Smelting Process 作者：刘丽丽 ; 左继红 ; 刘建华 英文作者：LIU Li-li;ZUO Ji-hong;LIU Jian-hua;College of Hunan Railway Professional Technology;Industrial University of Hunan; 关键词：铜闪速熔炼 ; 投影寻踪回归子模型 ; 预测输出 英文关键词：copper flash smelting;;projection pursuit regression model;;prediction output 中文刊名：ZDHJ 英文刊名：Techniques of Automation and Applications 机构：湖南铁道职业技术学院;湖南工业大学; 出版日期：2018-09-25 出版单位：自动化技术与应用 年：2018 期：v.37;No.279 基金：2015年度湖南省教育厅科学研究资助项目(编号15C0903) 语种：中文; 页：ZDHJ201809005 页数：4 CN：09 ISSN：23-1474/TP 分类号：24-27

摘要

提取金属铜的主要方式是闪速熔炼,由于铜矿资源复杂多变,为保证熔炼效果铜闪速熔炼过程需适应工况。目前铜闪速熔炼过程较难在线检测出冰铜温度、冰铜品位和渣中铁硅比,针对这个难题,提出了基于投影寻踪回归理论的铜闪速熔炼过程三大指标预测策略。首先构建三大工艺指标的投影寻踪回归子模型。并依据实数编码混沌伪并行遗传算法理论优化求解子模型参数。通过实验数据验证模型的预测误差比较小,能够满足实际生产需要。
        The main way to extract copper metal is flash smelting. Because of the complexity and variety of copper resources, copper flash smelting process needs to adapt the working conditions to ensure the smelting effect. The copper flash smelting process is difficult to measure the matte temperature, matte grade and ratio of iron and silicon in slag. For this problem, the three indexes prediction strategy of copper flash smelting process for projection pursuit regression theory based is put forward. First, the projection pursuit regression model of three major technological indexes is constructed. Based on the real coded chaotic pseudo parallel genetic algorithm, the sub model parameters are optimized. The experimental data show that the prediction error of the model is small and can meet the needs of actual production.

引文

[1]SANCHEZ-CORRALES V M,VALERA-GONZALEZJ A,PEREZ-TELLO M,et.al.Mass balance calculations in copper flash smelting by means of genetic algorithms[J].JOM Journal of the Minerals,Metals and Materials Society,2004,56(12):29-32.
    [2]FAGERLUND K O,JALKANEN H.Microscale simulation of settler processes in copper matte smel ting[J].Metallurgical and Materials Transactions B,2000,31(6):439-451.
    [3]LI X F,MEI C,XIAO T Y.Numerical simulation analysis of Guixi copper flash smelting furnace[J].Rare Metals,2002,21(4):260-265.
    [4]SCHAAF M,GóMEZ Z,CIPRIANO A.Real-time hybrid predictive modeling of the Teniente Converter[J].Journal of Process Control,2010,20(1):3-17.
    [5]HWANG T N,LAY S R,MAECHLER M.Re gression modelling in back-propagation and projection pursuit learning[J].IEEE Transaction on Neural Networks,1994,5(3):342-353.
    [6]赵恒平,俞金寿.化工数据预处理及其在建模中的应用[J].华东理工大学学报:自然科学版,2005,31(2):223-226.
    [7]QI X N,LIU Z Y,LI D D.Prediction of the per formance of a shower cooling tower based on projection pursuit regression[J].Applied Thermal Engineering,2008,28(8/9):1031-1038.
    [8]马森林.投影寻踪回归模型在林分出材率预测中的应用研究[D].福州:福建农林大学,2010.
    [9]马森林.投影寻踪回归模型在林分出材率预测中的应用研究[D].福州:福建农林大学,2010.
    [10]周东华,李钢,李元.数据驱动的工业过程故障诊断技术:基于主元分析与偏最小二乘的方法[M].北京:科学出版社,2011.
    [11]胡志坤,桂卫华,阳春华等.铜转炉吹炼过程熔剂加入量的模糊操作模式挖掘方法[J].控制与决策,2010,25(11):1689-1692.