欢迎访问《金属矿山》杂志官方网站,今天是 分享到:
×

扫码分享

金属矿山 ›› 2010, Vol. 39 ›› Issue (02): 26-29.

• 采矿工程 • 上一篇    下一篇

截止品位与入选品位的遗传-神经优化模型

刘婷,诸克军,贺勇,李玥   

  1. 中国地质大学
  • 出版日期:2010-02-15 发布日期:2010-11-09
  • 基金资助:

    国家自然科学基金项目(编号:70573101),武钢科研课题(编号:070429)。

An Integrating GA and ANN Model for Optimizing Cut-off Grade and Grade of Crude Ore

 LIU  Ting, ZHU  Ke-Jun, HE  Yong, LI  Yue   

  1. China University of Geosciences
  • Online:2010-02-15 Published:2010-11-09

摘要: 合理确定截止品位与入选品位关系到企业的经济效益和资源的可续性利用,有着重要的理论意义与现实意义。采用遗传算法与神经网络嵌套的动态优化模型,以武钢集团大冶铁矿为例,对其截止品位和入选品位进行动态优化。结果表明,当前大冶铁矿截止品位18%,入选品位41%~42%的生产方案有待改进,当截止品位为15.8%,入选品位取值为43.776%~44.139%,2007年1月—2007年11月的总净现值增加901~944万元。

关键词: 截止品位, 入选品位, 遗传算法, 神经网络

Abstract: The cut-off grade and grade of crude ore are decision-making variables which are directly related to the economic efficiency of enterprises and the utilization of renewable resources.Determining these two variables has a very important theoretical meaning and obvious practical significance.Taking Daye Iron Mine as an example, the dynamic optimized model of the GA and ANN was adopted to optimize the cut-off grade and grade of crude ore.The result showed that during the period of January to November in 2007, when the optimal cut-off grade was changed from 18% to 15.8%, and optimal grade of crude ore was from 41%~42% to 43.776%~44.139%, the net present value was improved by 9.01~9.44 million Yuan.

Key words: Cut-off grade, Grade of crude ore, Genetic algorithm, Neural networks