【作者】 刘业峰； 余刚； 郑秉霖； 柴天佑；

【Author】 Yefeng Liu~1,Gang Yu~1,Binglin Zheng~1,Tianyou Chai~(1,2) 1.Key Laboratory of Process Industry Automation,Ministry of Education,Northeastern University,Shenyang 110004,China 2.Research Center of Automation,Northeastern University,Shenyang 110004,China

【机构】 东北大学流程工业综合自动化教育部重点实验室； 东北大学自动化研究中心；

【摘要】 本文提出了一种基于PCA-GA-BP的神经网络模型,以判断综合生产指标经计划层层分解,下达各工序、作业班后的实时完成情况,进而对生产计划进行合理的修正。主成分分析(PCA)用来提取过程特征参数,剔除相关冗余信息;BP神经网络用来逼近非线性过程;改进了遗传算法(GA)的适应度函数,并对BP网络的权值和阀值进行确定。基于实际数据对选矿弱磁工序生产指标:弱磁精矿品位和弱磁尾矿品位进行预报,仿真结果表明所给出的模型对弱磁工序生产指标的预报是可靠的、准确的。更多还原

【Abstract】 In order to determine the global production indices’ real-time completion situation after plan’s layer upon layer’s decomposition and transmition to working procedure and work team.A neural network model based on PCA-GA-BP was proposed to reasonable modify the production plan.The principle component analysis(PCA) was used to select the most relevant process features and to eliminate the correlations of the input variables;back-propagation(BP) neural network was used to characterize the nonlinearity and accuracy;genetic algorithm(GA) was employed to optimize the parameters and structure of the BP neural network by improving GA’ fitness function.Carried on prediction to weak magnetic concentrate taste and weak magnetic tailings taste according to actual production data.The Simulation results show that the proposed method provides promising prediction reliability and accuracy.更多还原