磨矿分级系统溢流浓度的模糊智能控制

Fuzzy Intelligent Controlfor Overflow Density of Milling Classification Operation System

提出一种二维模糊智能控制器的算法，辅以基于规则的自学习环节，对具有诸多不确定因素、非线性、时变、难以建立精确数学模型的磨矿系统的溢流浓度进行控制．建造工艺专家系统，借助工业局域网络，综合不同性质的原矿石的要求和工艺专家知识，优化控制系统的给定参数．整个控制系统有较好的鲁棒性和稳定性，经过检测实际生产性能指标，溢流浓度的稳态误差不超过２－０ ％ ，且入磨矿石处理量提高了５－７ ％ ．

Algorithm oftwo dimension fuzzy intelligentcontroller andthe rule based self learning sys tem are used to controlthe overflow density of milling classification system . Uncertainty factors and non linear,time variation , however, prevail;they introduce difficulties to mathematical modelling . By applying the milling expertsystem with mineralproperty the controlled system’ssetting parametersmay be optimized in addition to improving the system stability . Performance of industrial process is improved by 5-7 % while stable state error of overflow densityisless than 2-0 % .