最优加权融合SO_2转化率预测模型  

Optimal weighted fusion SO_2 conversion ratio predication model

作　　者：王宁宁 

机构地区：[1]阿坝师范学院党委办公室,四川汶川623002

出　　处：《郑州轻工业学院学报（自然科学版）》2015年第3期138-141,共4页Journal of Zhengzhou University of Light Industry:Natural Science

基　　金：四川省教育厅科研项目(14ZB0339)

摘　　要：以冶炼烟气制酸SO2转化率为研究对象,针对单一BP或RBF神经网络,预测SO2转化率存在的过学习或网络速度收敛慢的问题,利用最优均方误差加权融合算法对两种单一神经网络进行融合,从而构建更优的SO2转化率预测模型.仿真结果表明:最优均方误差加权融合模型避免了单一模型信息的缺失,实现了信息互补,从而提高了SO2转化率预测精度.Selecting SO2 conversion ratio of metallurgical off-gases making acid as the research object,in order to solve the problem of over learning or slow network convergence speed existed in using single BP or RBF neural network to predict SO2 conversion ratio,the optimal mean square error weighted fusion algorithm was used to achieve fusion of two kinds of neural network and construct the better SO2 conversion ratio predication model. The simulation results showed that the optimal mean square error weighted fusion model avoided the lack of information of the single model,realized the complementation of information and effectively improved the SO2 conversion ratio predication precision.

关 键 词：神经网络 冶炼烟气制酸 SO2转化率 信息融合 最优加权融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TQ111[自动化与计算机技术—计算机科学与技术]