基于综合精矿回收率的配矿建模研究

The Research of Ore Blending Model Based on Comprehensive Recovery

配矿工作是选矿厂日常工作中的重要一环,合格的配矿产品能够使选矿工作更加顺利的进行。现有的配矿建模方式通常主要考虑配矿产品的质量而忽视配矿结果对选矿产品的影响。本文提出基于综合精矿回收率的配矿建模方法,通过BP神经网络算法拟合出配矿产品各项性质与综合精矿回收率之间的关系,并以此作为优化目标,从而在源头上保证选矿产品的质量。

Ore blending is an important part in the daily work of ore dressing plant. Reasonable blending can make the dressing work more smoothly. The existing model usually considers the quality of ore and ignores the impact on the dressing product. This paper presents an ore blending model based on comprehensive recovery. This model fits the relationship between the ores nature and the comprehensive recovery by BP neural network algorithm, then use the result as the optimization goal. This model can ensure the quality of dressing products from the source.