摘要
磨矿过程的粒度是直接关系到选矿生产精矿品位和金属回收率的重要指标,粒度的在线检测对磨矿过程的优化控制、提高精矿品位和金属回收率具有重要意义。但是,现有的粒度测量仪表检测周期长、价格昂贵、维护困难,难以实现在线测量。本文结合典型两段磨矿回路的特点,采用多输入层神经网络和遗传算法相结合的方法,提出了采用实数编码遗传算法训练多输入层神经网络的混合算法,建立了磨矿粒度的神经网络软测量模型,并通过现场数据验证和实际应用验证了本文方法的有效性。
The particle size of grinding circuit is the important performance index directly related to the grade of concentrated ore and metal recovery rate. The measurement of the particle size is the key to realize the optimizing control of the grinding circuit and to improve the grade of concentrated ore and metal recovery rate. The problem of the present instrument is that the particle size can＇t be measured on line due to the long measurement period, expensive price and maintenance difficulty. Based on the characteristics of the typical two stage grinding circuit, a hybrid algorithm is proposed, which adopts multi-layer input neural network （NN） and real-coded genetic algorithm. The NN soft-sensor model of particle size is established through training the NN with the genetic algorithm. The field test data and the application result show that the proposed soft-sensor approach is efficient.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2006年第9期981-984,共4页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(60534010)
国家973重点基础研究计划项目(2002CB312201)
新世纪优秀人才支持计划(NCET-05-0294)资助项目
关键词
磨矿过程
磨矿粒度
软测量
遗传算法
grinding circuit particle size soft-sensor genetic algorithm
