摘要
为提高露天矿生产运输效率,实现生产资源的有效回收和低成本运营,提出基于神经网络代理辅助的多目标露天矿卡车优化调度方法。依据露天矿卡车生产运输的特点,综合考虑卡车运输的各项成本,构建以卡车运输成本最小、卡车总等待时间最短,以及卸矿站品位偏差率最小为目标的多目标组合优化模型。在基本人工鱼群算法基础上,通过引入神经网络代理辅助模型训练历史调度数据,对雄安新区某露天矿的生产调度情况进行了仿真研究。结果表明,基于神经网络代理辅助的人工鱼群算法可有效提高算法计算速度约47.8%,降低露天矿卡车运输成本8.16%,减少等待时间12.12%,并减小品位偏差8.69%,在加快计算速度的同时,实现了平衡算法收敛性与多样性的目标。
In order to increase production and transportation efficiency of open-pit mine,and to realize the effective recovery of resources and low cost operation,a multi-objective open-pit truck optimal scheduling technique based on neural network agent assistance was proposed.According to the transportation characteristic of open-pit truck,considering all the costs of truck dispatching and transportation,a multi-objective combinatorial optimization model with the objectives of minimizing truck transportation cost,minimizing total waiting time of truck and minimizing grade deviation rate of unloading station was constructed.Based on the basic artificial fish swarm algorithm,the production scheduling of an open-pit mine in Xiong’an new area was simulated by introducing the neural network agent-aided model to train the historical scheduling data.The results show that the artificial fish swarm algorithm based on neural network agent assistance can effectively improve the calculation speed of the algorithm by about 47.8%,reduce the truck transportation cost of open-pit mine by 8.16%,reduce the waiting time by 12.12%,and reduce the grade deviation by 8.69%.While accelerating the calculation speed,the goal of balancing the convergence and diversity of the algorithm has been realized.
作者
阮顺领
金裕
李发本
顾清华
王丹娜
RUAN Shunling;JIN Yu;LI Faben;GU Qinghua;WANG Danna(School of Resources Engineering,Xi'an University of Architecture and Technology,Xi'an,Shaanxi 710055,China;Luoyang Molybdenum Company and Limited,Luoyang,Henan 471500,China;School of Management,Xi’an University of Architecture and Technology,Xi’an,Shaanxi 710055,China)
出处
《矿业研究与开发》
CAS
北大核心
2021年第8期154-160,共7页
Mining Research and Development
基金
国家自然科学面上项目(52074205)
陕西省自然科学基础研究计划联合基金重点项目(2019JLP-16)
陕西省自然科学基础研究计划杰青项目(2020JC-44)。
关键词
露天矿
卡车调度
优化调度
神经网络代理辅助
人工鱼群
Open-pit mine
Truck scheduling
Optimal scheduling
Neural network agent assistance
Artificial fish swarm
