基于DE-GWO算法的矿井风网风量调节方法  

Method for adjusting air volume of mine ventilation network based on DE-GWO algorithm

作　　者：吴新忠 张芝超 王凯[2] 韩正化 魏连江[2] WU Xinzhong;ZHANG Zhichao;WANG Kai;HAN Zhenghua;WEI Lianjiang(School of Information and Electrical Engineering,China University of Mining&Technology,Xuzhou 221116,China;School of Safety Engineering,China University of Mining&Technology,Xuzhou 221116,China)

机构地区：[1]中国矿业大学信息与控制工程学院,江苏徐州221116 [2]中国矿业大学安全工程学院,江苏徐州221116

出　　处：《中南大学学报（自然科学版）》2021年第11期3981-3989,共9页Journal of Central South University:Science and Technology

基　　金：国家重点研发计划项目(2018YFC0808100);江苏省重点研发计划项目(BE2016046)。

摘　　要：为及时满足井下某一用风地点的需风要求,提出一种根据分支风量期望值来选择不同调节分支的智能应急调风方案。首先,以矿井通风网络需风分支风量可调最大化为目标,建立通风网络的非线性优化数学模型,针对该优化模型中的风量平衡、风压平衡等约束条件,采用不可微精确罚函数将其转化为目标模型中的惩罚项。然后,通过对风网灵敏度矩阵的计算求解,得出最优的可调分支集和风阻调节范围,并基于灰狼算法实现寻优。为克服灰狼算法在求解复杂问题上易陷入局部最优的缺陷,提出一种差分灰狼算法(DE-GWO),该方法在对种群个体位置不断迭代更新的基础上,加入差分进化算法的变异、交叉、选择操作,进而维持种群的多样性。最后,基于矿井智能控制系统实验平台验证应急调风方案的可行性。研究结果表明:相比于GWO算法,本文提出的DE-GWO算法在寻优性能和稳定性方面提升显著,可用于及时调节风量。In order to meet the wind demand of a certain underground location in time,an intelligent emergency air regulation scheme that selects different branches according to the expected value of branch air volume was proposed.Firstly,the nonlinear optimization mathematical model of ventilation network was established with the goal of maximizing the adjustable air volume of the branch of mine ventilation network.Aiming at the constraint conditions of air volume balance and air pressure balance in the optimization model,the non differentiable exact penalty function was used to transform it into the penalty term in the target model.Then,the optimal adjustable branch set and wind resistance regulation range were obtained by calculating the wind network sensitivity matrix,and the optimization was realized based on gray wolf algorithm.In order to make up for the defect that the gray wolf algorithm was easy to fall into local optimum in solving complex problems,a differential evolution gray wolf optimizer(DE-GWO)was proposed,which was mainly based on the continuous iterative updating of individual position of population by gray wolf algorithm,adding mutation,crossover and selection operations of differential evolution algorithm,so as to maintain the diversity of population.Finally,based on the experimental platform of mine intelligent control system,the feasibility of emergency air conditioning scheme was verified.The results show that compared with GWO algorithm,the DE-GWO algorithm proposed in this paper has a significant improvement in the optimization performance and stability,and can be used to adjust the air volume in time.

关 键 词：矿井通风 灵敏度 差分灰狼算法(DE-GWO) 应急调风 

分 类 号：TD724[矿业工程—矿井通风与安全]