系统仿真学报 ›› 2021, Vol. 33 ›› Issue (9): 2085-2094.doi: 10.16182/j.issn1004731x.joss.20-0403

• 仿真建模理论与方法 • 上一篇    下一篇

基于模拟退火与改进粒子群的矿井通风优化算法

邵良杉1,2, 王振1,2, 李昌明1,2   

  1. 1.辽宁工程技术大学 系统工程研究所,辽宁 葫芦岛 125105;
    2.辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105
  • 收稿日期:2020-06-23 修回日期:2020-08-24 出版日期:2021-09-18 发布日期:2021-09-17
  • 作者简介:邵良杉(1961-),男,博士,博士生导师,教授,研究方向为矿业系统工程。E-mail:Intushao@163.com
  • 基金资助:
    国家自然科学基金(71771111)

Optimization Algorithm of Mine Ventilation Based on SA-IPSO

Shao Liangshan1,2, Wang Zhen1,2, Li Changming1,2   

  1. 1. Liaoning Technical University Institute of Systems Engineering, Huludao 125105, China;
    2. School of Software, Liaoning Technical University, Huludao, 125105, China
  • Received:2020-06-23 Revised:2020-08-24 Online:2021-09-18 Published:2021-09-17

摘要: 建立以矿井通风网络总功率最小为目标的非线性优化数学模型,改进粒子群优化算法实现寻优。在粒子群算法中引入变异操作,提出了一种新的惯性权重,并构造一种新的粒子选择方法控制违反约束条件的粒子数量,提高粒子群算法寻找边界的能力。将风量平衡定律和风压平衡定律的约束条件转化为目标函数的惩罚项,改进粒子群(Improved Particle Swarm Optimization, IPSO)对目标函数进行优化,利用模拟退火在搜索过程中引入随机因素达到全局最优。对新屯矿系统进行仿真模拟实验,结果表明:该算法能够降低通风总能耗的95.69 kW,且风量值满足用风需求。

关键词: 变异操作, 惯性权重, 约束问题, 改进粒子群, 惩罚项, 模拟退火

Abstract: A non-linear optimization mathematical model aiming at the minimum total power of the mine ventilation network is established, in which SA-IPSO algorithm is applied for the optimization. The mutation operation is introduced in the PSO algorithm, in which a new inertia weight is proposed and a new particle selection method is constructed to control the number of particles violating the constraints, and the ability of the PSO algorithm to find boundaries is improved. The constraint conditions of the mine ventilation law are transformed into the penalty term of the objective function. IPSO optimizes the objective function and uses SA to achieve the global optimum. The simulation experiment of Xintun Mine shows that the algorithm can reduce the total energy consumption of ventilation by 95.69Kw while the air volume can meet the demand.

Key words: mutation operation, inertia weight, constraint problem, improved particle swarm, penalty term, simulated annealing

中图分类号: