|
|||||
|
|
| 品位组合优化的粒子群—神经计算方法 | |
| 引用本文： | 贺勇,诸克军,胡承凡. 品位组合优化的粒子群—神经计算方法[J]. 系统管理学报, 2010, 19(2) |
| 作者姓名： | 贺勇 诸克军 胡承凡 |
| 作者单位： | 1. 广东工业大学,管理学院,广州,510520 2. 中国地质大学,经济管理学院,武汉,430074 3. 武汉钢铁集团矿业有限责任公司,大冶铁矿,湖北,黄石,435006 |
| 基金项目： | 国家自然科学基金,武钢科研课题,广东工业大学博士启动项目 |
| 摘 要： | 根据矿山采选物理过程,建立以截止品位与入选品位为决策变量,经济效益为目标函数的非线性模型,并采用粒子群-神经集成的方法进行优化求解.其基本操作为:截止品位与入选品位组合构成进化计算的个体,个体包括两部分,前半部分代表截止品位,后半部分代表入选品位;用BP网络和RBF网络建立收益(适应度函数)与粒子个体的局部联系;利用粒子群算法的全局搜索功能找出使适应度函数最大时的品位组合(截止品位及入选品位).研究表明: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17.833 7%～17.836 7%,入选品位取值为46.4%,2007-01～2007-11的精矿量增加13.92万t,总净现值增加6.698百万元. |
| 关 键 词： | 品位优化 粒子群算法 神经网络 截止品位 入选品位 |
A Method of Integrating PSO and ANN for Optimizing Cut-off Grade and Grade of Crude Ore |
|
| HE Yong,ZHU Ke-jun,HU Cheng-fan. A Method of Integrating PSO and ANN for Optimizing Cut-off Grade and Grade of Crude Ore[J]. Systems Engineering Theory·Methodology·Applications, 2010, 19(2) | |
| Authors: | HE Yong ZHU Ke-jun HU Cheng-fan |
| Affiliation: | HE Yong~1,ZHU Ke-jun~2,HU Cheng-fan~3(1.School of Management,Guangdong University of Technology,Guangzhou 510520,China,2.School of Economic , Management,China University of Geosciences,Wuhan 430074,3.Daye Iron Mine,Wuhan Iron , Steel Group Co.Ltd.,Huangshi 435006,China) |
| Abstract: | According to the mining & milling production of metal mine,this paper establishes a nonlinear model,in which cut-off grade and grade of crude ore are decision-making variables,economic benefit is on(objective) function,and integrates PSO and ANN to optimize the Cut-off grade and Grade of crude ore.The(idea) is detailed as follows: put Cut-off grade and Grade of crude ore together as particles for evolution computation,a particle stands for a combination of cut-off grade and grade of crude ore;use BPNN and R... |
| Keywords: | Grade optimization PSO algorithm neural networks cut-off grade grade of crude ore |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |