粗集神经网络在阳离子反浮选控制中的应用  

Rough Set-Neural Network and its Application to Cation Anti-flotation Control

作　　者：张勇 丛峰武 王伟 

机构地区：[1]大连理工大学信息与控制研究中心 [2]鞍钢弓长岭矿业公司,辽宁辽阳111007

出　　处：《控制工程》2004年第S2期85-88,92,共5页Control Engineering of China

基　　金：教育部科学技术研究重点资助项目(01053)辽宁省教育厅科研基金资助项目(202193396)

摘　　要：针对阳离子反浮选生产过程被控对象复杂、数学模型不确定以及控制精度要求高等特点,提出一种基于粗集神经网络理论的智能控制模型。该方法减少信息表达的属性数量及神经网络构成系统的复杂性,增强了系统容错及抗干扰的能力。将粗集神经网络智能控制模型与基于粗集的控制模型进行对比,结果表明了该方法的可行性。An intelligent control method based on the rough set-neural network theory is proposed for a cation anti-flotation process due to its complexity, model uncertainty and the high requirement of its control quality. This method reduces the quantity of attributes and the complexity of the system made up of neural network, and improves the ability of containing mistake and anti-disturbance. The simulation comparing rough set model with rough set-neural network model show the feasibility of the method suggested.

关 键 词：浮选控制 粗集理论 神经网络 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]