节点文献
改进的RBF神经网络在磨矿指标预测中的应用
Application of Improved RBF Neural Network to the Prediction of Grinding Production Index
【摘要】 针对选矿生产过程中磨矿生产率和磨矿浓度、介质填充率和料球比3个指标之间的关系,分析了磨矿生产率对磨矿过程产品质量和工作效率的影响,提出了磨矿过程预测模型,采用了模糊聚类与改进的遗传算法相结合训练RBF神经网络的新方法。仿真结果表明,该模型可以快速准确地预报生产状况,对于企业进行实时生产控制具有较高的实用价值。
【Abstract】 Aiming at the relationship between the grinding production rate and the three indexes of grinding concentration,charge ratio of media and ration of powder to ball in the mineral processing,the influence of the grinding production rate to the quality of product in grinding processing and working efficiency is analyzed.And a grinding processing predicting model is proposed.A new method which combines the fuzzy gathering and genetic algorithm based on the fuzzy logic is used to train the RBF neural network.The simulation results show that this model can predict grinding production condition quickly and precisely,and has a higher practical value for the enterprise to carry on the real-time production control.
【Key words】 grinding production rate; fuzzy clustering; RBF neural networks; genetic algorithm; predicting model;
- 【文献出处】 控制工程 ,Control Engineering of China , 编辑部邮箱 ,2008年05期
- 【分类号】TP183;TD921.4
- 【被引频次】6
- 【下载频次】170