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 ›› 2026, Vol. 55 ›› Issue (9): 1-12.

• 《金属矿山》创刊60周年成果专栏 •    下一篇

面向回采装备多目标调度的改进遗传退火算法

葛世荣1,2,3,郝雪弟1,2,何福强1,杨 建1,王致远1,鲁自横1   

  1. 1.中国矿业大学(北京)机械与电气工程学院,北京 100083;2.中国矿业大学(北京)江西研究院,江西 赣州 341000; 3.江西理工大学,江西 赣州 341000
  • 出版日期:2026-09-15 发布日期:2026-10-13
  • 通讯作者: 郝雪弟(1974—),男,副教授,博士,硕士研究生导师。
  • 作者简介:葛世荣(1963—),男,中国工程院院士,教授,博士,博士研究生导师。
  • 基金资助:
    “十四五”国家重点研发计划项目(编号:2022YFC2904105)。

Improved Genetic Simulated Annealing Algorithm for Multi-objective Scheduling of Extraction Equipment

GE Shirong1,2,3,HAO Xuedi1,2 ,HE Fuqiang1 ,YANG Jian1 ,WANG Zhiyuan1 ,LU Ziheng1   

  1. 1.School of Mechanical and Electrical Engineering,China University of Mining and Technology-Beijing,Beijing 100083; 2.Jiangxi Research Institute of China University of Mining and Technology-Beijing,Ganzhou 341000,China; 3.Jiangxi University of Science and Technology,Ganzhou 341000,China
  • Online:2026-09-15 Published:2026-10-13

摘要: 针对地下金属矿山传统单目标调度模型适应性不足的问题,以总完工时间与总工序间隔时间最小化为 优化目标,结合矿石开采、运输、提升等环节中装备运行状态与资源需求的实际工况特征,构建了回采装备多目标调 度优化模型。为将多目标调度问题合理转化为可求解的单目标决策模型,设计了一种基于层次分析法-熵权法(Ana lytic Hierarchy Process-Entropy Weight Method,AHP-EWM)的多目标综合赋权方法,通过静态赋权与动态权重调整相结 合的策略,实现多目标参数的精确权重配置。以赋权后的单目标综合优化模型为求解对象,针对传统遗传算法收敛 速度慢、易陷于局部最优的固有缺陷,引入自适应算子与模拟退火机制,提出了改进遗传退火算法(Improved Genetic Simulated Annealing Algorithm,IGASA)。实例验证结果表明:相较于传统调度方法,所提方法在总完工时间上降低 36.54%,总工序间隔时间减少38.13%,调度性能提升显著。进一步地,为将所求解的调度方案直观应用于生产管理, 提出了生产调度数字孪生系统的流程架构,支持调度计划的动态仿真与实时监控,从而形成“权重配置—模型求解— 可视化呈现”的完整技术链条,为地下矿山智能化调度系统的构建提供了理论依据与技术支撑。

关键词: 多目标调度, 改进遗传退火算法, 熵权法, 地下金属矿山, 数字孪生, 回采装备, 权重配置

Abstract: To address the insufficient adaptability of traditional single-objective scheduling models for underground metal mines,a multi-objective scheduling optimization model for stoping equipment was constructed.The model takes the minimiza tion of the makespan and total inter-operation interval time as the optimization objectives and incorporates the actual working condition characteristics of equipment operating states and resource requirements in ore mining,haulage,and hoisting.To rea sonably transform the multi-objective scheduling problem into a solvable single-objective decision model,a multi-objective com prehensive weighting method based on the Analytic Hierarchy Process-Entropy Weight Method (AHP-EWM) was designed.By combining static weighting with dynamic weight adjustment,this method achieves precise weight configuration of multi-objective parameters.Taking the weighted single-objective comprehensive optimization model as the object to be solved,and considering the inherent shortcomings of traditional genetic algorithms,such as slow convergence and susceptibility to local optima,adaptive operators and a simulated annealing mechanism were introduced,and an Improved Genetic Simulated Annealing Algorithm (IGASA) was proposed.Case study results show that,compared with conventional scheduling methods,the proposed method reduces the makespan by 36.54% and the total inter-operation interval time by 38.13%,indicating significantly improved scheduling performance.Furthermore,to intuitively apply the obtained scheduling scheme to production management,a process architecture for a production scheduling digital twin system was proposed,which supports dynamic simulation and real-time mo nitoring of scheduling plans,thereby forming a complete technical chain of "weight configuration-model solving-visualization presentation" and providing a theoretical basis and technical support for constructing intelligent scheduling systems for under ground mines.

Key words: multi-objective scheduling,IGASA,EWM,underground metal mine,digital twin,stoping equipment,weight configuration

中图分类号: