摘要
多金属配矿的目的是为了保证入选矿石品位的稳定和提高矿石的综合回收率。针对现有多金属配矿模型对选矿影响因素考虑不充分,导致精细化程度不足的问题,在考虑影响选矿回收率的矿石岩性、氧化率、有害物质等配矿指标的情况下,建立了以品位偏差、矿石岩性百分比偏差、总产量偏差和采掘运输成本为优化目标,基于采选流程的多金属多目标配矿优化数学模型。在标准遗传算法的基础上,对算法的变异过程和比较选择过程进行了改进,设计了一种自适应遗传算法对该模型进行求解。以国内某大型钼钨矿为例,将该模型的优化结果与传统模型的优化结果进行对比分析,并采用遗传算法、粒子群算法和所提出的自适应遗传算法分别进行模型求解。研究表明:①该模型在保证采掘运输成本不增加的情况下,对原有配矿模型未考虑的矿石岩性、氧化率、有害物质等影响因素进行了配矿优化,保证了矿石品位的均衡性以及矿石的可选性,从而增加了配矿优化模型的适用性,使得配矿优化模型更符合生产实际;②所提出的自适应遗传算法不仅能够避免陷入局部最优解,而且在计算效率上相比于粒子群算法和遗传算法分别提高了40%和56%,在模型求解精度上提高了10倍左右,表明了改进算法的有效性。
The purpose of polymetallic ore blending is to ensure the stability of the selected ore grade and to improve the overall recovery rate of the ore.Aiming at the problem that the existing polymetallic ore blending model has insufficient refinement due to insufficient consideration of factors affecting mineral processing,and under the comprehensive consideration of the indicators such as ore lithology,oxidation rate and hazardous materials that affects the ore recovery rate,a multi-metal multi-objective ore blending model based on grade deviation,ore lithology percentage deviation,total production deviation and mining transportation cost was established.By adapting the variation process and selection process of the standard genetic algorithm,an adaptive genetic algorithm is designed to solve the model.Taking a large molybdenum-tungsten mine in China as a case,the optimization results of this model are compared with the traditional models,and the genetic algorithm,particle swarm optimization algorithm and the proposed adaptive genetic algorithm are used to solve the model.The study results show that:①this model optimizes the ore influencing factors such as ore lithology,oxidation rate and harmful substances that could not be considered in the case of ensuring that the transportation cost is not increased,and it ensures the balance of the ore grade and the optionality of the ore,thereby increasing the applicability of the ore optimization model,making the ore optimization model more in line with production practice;②the proposed adaptive genetic algorithm not only avoids falling into the local optimal solution,but also increases the computational efficiency by 40%and 56%,compared with the particle swarm algorithm and the genetic algorithm respectively,and the solution accuracy is improved by 10 times,which proves the effectiveness of the improved algorithm.
作者
顾清华
刘海龙
卢才武
李学现
杨亚鹏
Gu Qinghua;Liu Hailong;Lu Caiwu;Li Xuexian;Yang Yapeng(School of Management,Xi’an University of Architecture and Technology,Xi’an 710055,China;School of Resources Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China;Luomo Group Mineral Processing Company,Luoyang 471500,China)
出处
《金属矿山》
CAS
北大核心
2020年第3期56-63,共8页
Metal Mine
基金
国家自然科学基金项目(编号:51774228)
陕西省自然科学基金项目(编号:2017JM5043)
陕西省教育厅专项科研计划项目(编号:17JK0425)。
关键词
露天矿
配矿优化模型
自适应遗传算法
多目标优化
Open-pit mine
Ore blending optimization model
Adaptive genetic algorithm
Multi-objective optimization
