| 441 | 9 | 690 |
| 下载次数 | 被引频次 | 阅读次数 |
为了改善传统的基于多目标优化算法的露天矿车辆调度存在的实时性差和能耗比低的问题，提出一种基于多智能体强化学习的露天矿车辆调度方法。通过对露天矿场车辆、装载点及卸载点等相关信息的分析，构建多智能体强化学习算法所需的仿真环境。同时，基于集中式训练分布式执行范式构建一个基于社交感知的多智能体深度确定性策略梯度模型，将其作为强化学习的智能体，该模型的演员网络通过使用深度确定性策略梯度算法保证所提调度方法的稳定性。仿真结果表明，所提方法能满足露天矿车辆调度的实时调度，并且能提高能耗比。
Abstract：In order to improve the real-time performance and low fuel consumption rate of traditional open-pit mine vehicle scheduling based on multi-objective optimization algorithms, a multi-agent reinforcement learning-based open-pit mine vehicle scheduling method is proposed.By analyzing relevant information such as vehicles, loading points and unloading points in open-pit mines, a simulation environment required for multi-agent reinforcement learning algorithms is constructed.At the same time, based on the centralized training and distributed execution paradigm, a socially aware multi-agent deep deterministic policy gradient model is constructed as the intelligent agent of reinforcement learning.The actor network of this model used the deep deterministic policy gradient algorithm to ensure the stability of the proposed scheduling algorithm.Simulation results show that the proposed method can meet the requirements of low fuel consumption rate and real-time performance of open-pit mine vehicle scheduling.
[1] 付恩三，刘光伟，邸帅，等.露天矿山无人驾驶技术及系统架构研究[J].煤炭工程，2022,54(1):34-39.FU E S,LIU G W,DI S,et al.Unmanned driving technology and system architecture in open-pit mines[J].Coal Engineering,2022,54(1):34-39.(in Chinese)
[2] 王富民，贺昌斌.露天矿卡车无人驾驶技术的现状与展望[J].露天采矿技术，2021,36(3):45-47.WANG F,HE C B.Status and outlook of autonomous driving technology for trucks in open-pit mine[J].Opencast Mining Technology,2021,36(3):45-47.(in Chinese)
[3] 武讲，郑群飞.哈尔乌素露天矿无人驾驶方案研究[J].金属矿山，2021,50(2):167-172.WU J,ZHENG Q F.Feasibility study of driverless system in Harwusu Open-pit Coal Mine[J].Metal Mine,2021,50(2):167-172.(in Chinese)
[4] 张媛.综合成本最小的露天矿卡车低碳调度优化研究[D].西安：西安建筑科技大学，2020.ZHANG Y.Research on optimization of low carbon truck dispatching in open pit mines with the least comprehensive Cost [D].Xi'an:Xi'an University of Architecture and Technology,2020.(in Chinese)
[5] 郭天中.基于SA-GA算法的露天矿用卡车调度方法的研究[D].包头：内蒙古科技大学，2015.GUO T Z.Research on the scheduling method of open-pit mining trucks based on SA-GA algorithm [D].Baotou:Inner Mongolia University of Science and Technology,2015.(in Chinese)
[6] 安江川.基于GIS的虚拟露天采矿系统研究[D].西安：西安建筑科技大学，2008.AN J C.Research on virtual open pit mining system based on GIS [D].Xi'an:Xi'an University of Architecture and Technology,2008.(in Chinese)
[7] 武讲，郑群飞.哈尔乌素露天矿无人驾驶方案研究[J].金属矿山，2021,50(2):167-172.WU J,ZHENG Q F.Research on unmanned driving plan for harwusu open pit mine[J].Metal Mines,2021,50 (2):167-172.
[8] 陈建宏，郑海力，刘浪，等.基于排队论与灰色关联的露天矿排土车辆配置优化[J].矿业研究与开发，2009,29(6):3-5.CHEN J H,ZHENG H,LIU L,et al.Optimization of dumping vehicle configuration in open pit mines based on queuing theory and grey correlation[J].Mining Research and Development,2009,29(6):3-5.(in Chinese)
[9] 王金亮.基于遗传算法的多车型露天矿卡车调度模型研究[J].现代矿业，2021,37(3):35-39.WANG J L.Research on a multi-vehicle open pit truck scheduling model based on genetic algorithm[J].Modern Mining,2021,37 (3):35-39.(in Chinese)
[10] 程平，李晓光，顾清华，等.露天矿新能源纯电动卡车的智能调度优化及应用[J].金属矿山，2023(3):193-198.CHENG P,LI X G,GU Q H,et al.Intelligent scheduling optimization and application of new energy pure electric trucks in open-pit mines[J].Metal Mines,2023(3):193-198.(in Chinese)
[11] 张明，顾清华，李发本，等.基于多目标遗传算法的露天矿卡车调度优化研究[J].金属矿山，2019(6):157-162.ZHANG M,GU Q H,LI F B,et al.Optimization of truck scheduling in open-pit mines based on multi-objective genetic algorithm[J].Metal Mines,2019(6):157-162.(in Chinese)
[12] 阮顺领，金裕，李发本，等.基于人工鱼群神经网络进化的露天矿卡车优化调度研究[J].矿业研究与开发，2021,41(8):154-160.RUAN S L,JIN Y,LI F B,et al.Optimization scheduling of open-pit mining trucks based on artificial fish swarm neural network evolution[J].Mining Research and Development,2021,41(8):154-160.(in Chinese)
[13] 叶芳泽，沈炜.基于深度强化学习改进的任务调度算法[J].计算机时代，2022(11):55-58.YE F Z,SHEN W.Improved task scheduling algorithm based on deep reinforcement learning[J].Computer Age,2022(11):55-58.(in Chinese)
[14] 贺俊杰，张洁，张朋，等.基于长短期记忆近端策略优化强化学习的等效并行机在线调度方法[J].中国机械工程，2022,33(3):329-338.HE J J,ZHANG J,ZHANG P,et al.Related parallel machine online scheduling method based on LSTM-PPO reinforcement learning[J].China Mechanical Engineering,2022,33(3):329-338(in Chinese)
[15] 张玉芬，朱雅琳.马尔可夫性及其检验方法研究[J].价值工程，2012,31(2):312-313.ZHNAG Y F,ZHU Y L.The study of Markov property and its test[J].Value Engineering,2012,31(2):312-313.(in Chinese)
[16] 高阳，陈世福，陆鑫.强化学习研究综述[J].自动化学报，2004,30(1):15-20.GAO Y,CHEN S F,LU X.A review of research on reinforcement learning[J].Journal of Automation,2004,30(1):15-20.(in Chinese)
[17] 李在友，孙艳斌，王晓光，等.基于改进被囊群算法的露天矿无人驾驶卡车运输调度[J].工矿自动化，2022,48(6):87-94.LI Z Y,SUN Y B,WANG X G,et al.Unmanned truck transportation scheduling in open-pit mines based on improved swarm optimization algorithm[J].Industrial and Mining Automation,2022,48(6):87-94.(in Chinese)
[18] LOWE R,WU Y,TAMAR A,et al.Multi-agent actor-critic for mixed cooperative-competitive environments[EB/OL].[2022-11-01].https://dl.acm.org/doi/10.5555/3295222.3295385.
基本信息:
DOI：10.13682/j.issn.2095-6533.2023.01.012
中图分类号:TD57;TP18
引用信息:
[1]李新华.基于多智能体强化学习的露天矿车辆调度方法[J].西安邮电大学学报,2023,28(01):104-110.DOI:10.13682/j.issn.2095-6533.2023.01.012.
2022-12-14
2022
2023-05-30
2024-02-04
2024
2
2023-01-10
2023-01-10