Research on intelligent control of air volume of mine ventilation network
-
摘要: 现有矿井通风网络风量智能优化算法在求解调风参数时普遍存在模型复杂、收敛速度慢、易陷入局部最优等缺陷，同时也缺乏与调风分支优化选择相结合的研究。针对上述问题，提出了一种基于改进天牛须搜索（BAS）算法的矿井通风网络风量智能调控方法。首先，以用风分支的风量需求为优化目标，构建风量优化调节数学模型，针对该模型中的风量调节约束条件，采用不可微精确罚函数并结合模拟退火算法优化惩罚项，实现模型的去约束化。然后，通过求解灵敏度矩阵，结合风量灵敏度和分支支配度理论选择最优的调节分支集，确定其风阻调节范围，并作为模型的初始解集。最后，基于改进BAS算法求解出最优调风参数，进而控制对应的调风设施，实现风量调控。基于矿井通风实验平台对该方法的可靠性进行实验验证，结果表明：相比于标准BAS算法和粒子群优化（PSO）算法，改进BAS算法综合寻优性能更优越，解得的风量平均值和最优解均高于PSO算法和标准BAS算法，平均运行时间虽略长于标准BAS算法，但远短于PSO算法，平均收敛代数最多，精度最高，容易跳出局部循环得到最优解；在设定风量调节目标后，基于改进BAS算法的矿井通风网络风量智能调控方法可快速精准求解出待调分支的风量最优值，调节后的分支风量满足矿井安全生产的调风要求，风量上调高达 46.5%。Abstract: The existing intelligent optimization algorithm of air volume of mine ventilation network has the defects of complex model, slow convergence speed, easy falling into local optimum when solving the air adjustment parameters. There is a lack of research on the combination of optimal selection of air adjustment branches. To solve the above problems, an intelligent control method of air volume of mine ventilation network based on improved beetle antennae search (BAS) algorithm is proposed. Firstly, the mathematical model of air volume optimal adjustment is established by taking the air volume demand of the air consumption branch as the optimization objective. In view of the air volume adjustment constraint conditions in the model, the non-differentiable exact penalty function and the simulated annealing algorithm are adopted to optimize the penalty term, so that the model is unconstrained. Secondly, by solving the sensitivity matrix and combining the theory of air volume sensitivity and branch dominance, the optimal adjustment branch set is selected. The air resistance adjustment range is determined as the initial solution set of the model. Finally, based on the improved BAS algorithm, the optimal air adjustment parameters are solved. The corresponding air adjustment facilities are controlled to realize air volume adjustment. The reliability of the method is verified by experiments based on the mine ventilation experimental platform. The results show that compared with the standard BAS algorithm and particle swarm optimization (PSO) algorithm, the improved BAS algorithm has superior comprehensive optimization performance. The average value and optimal solution of air volume are higher than those of the PSO algorithm and standard BAS algorithm. Although the average running time is slightly longer than the standard BAS algorithm, it is far shorter than the PSO algorithm. The average convergence algebra is the most, the precision is the highest, and it is easy to jump out of the local loop to get the optimal solution. After setting the air volume adjustment target, the intelligent control method of air volume of the mine ventilation network based on the improved BAS algorithm can quickly and accurately solve the optimal value of the air volume of the branch to be adjusted. The adjusted branch air volume meets the air volume adjustment requirements of mine safety production, and the air volume is increased up to 46.5%.
-
图 5 灵敏度
$ {d}_{\mathrm{4,18}} $ 随风阻$ {R}_{18} $ 的变化Figure 5. Variation of sensitivity
$ {d}_{\mathrm{4,18}} $ with air resistance$ {R}_{18} $ 
图 6 不同算法所得分支4风量适应度曲线
Figure 6. Air volume fitness curves of branch 4 obtained by different algorithms
图 7 分支4瓦斯体积分数随风量变化曲线
Figure 7. Change curves of gas volume fraction in branch 4 with air volume
表 1 通风网络初始参数
Table 1. Initial parameters of ventilation network
分支
编号始节点 末节点 风阻/
(N·s2·m−8)初始风量/
(m3 ·s−1)最小需风量/
(m3 ·s−1)分支
编号始节点 末节点 风阻/
(N·s2·m−8)初始风量/
(m3 ·s−1)最小需风量/
(m3 ·s−1)1 ① ② 0.455 43.37 41.92 12 ⑥ ⑩ 1.376 5.48 2.59 2 ② ③ 0.208 20.60 16.85 13 ⑦ ⑨ 1.206 5.43 5.46 3 ② ④ 0.124 22.78 19.43 14 ⑧ ⑩ 0.336 5.62 4.26 4 ③ ⑪ 1.156 9.57 6.65 15 ⑨ ⑪ 0.209 14.79 9.94 5 ③ ⑤ 0.197 11.02 7.29 16 ⑩ ⑬ 0.137 11.11 6.71 6 ④ ⑥ 0.040 14.88 11.26 17 ⑪ ⑫ 0.074 24.37 20.38 7 ④ ⑬ 1.076 7.90 6.14 18 ⑬ ⑫ 0.296 19.01 12.42 8 ⑤ ⑨ 0.415 9.36 6.07 19 ⑫ ⑭ 0.129 43.37 41.92 9 ⑦ ⑤ 0.327 1.66 1.24 20 ⑭ ⑮ 0.727 43.37 41.92 10 ⑧ ⑦ 0.647 3.77 2.12 21 ⑮ ① 0 43.37 41.92 11 ⑥ ⑧ 0.349 9.39 7.34 
下载: 导出CSV
表 2 分支4风量的灵敏度和支配度
Table 2. Sensitivity and dominance of the air volume of branch 4
分支编号 灵敏度 支配度 分支编号 灵敏度 支配度 1 3.2169 124.3317 12 0.0573 5.9822 2 3.5545 119.3313 13 0.1969 6.2548 3 3.4561 146.8510 14 0.1522 11.5349 4 2.8358 15.6208 15 4.2291 56.1725 5 2.7374 40.1354 16 0.8291 41.8228 6 0.7791 78.3699 17 6.8767 152.2138 7 0.1961 13.1203 18 3.5631 90.9727 8 1.1088 23.9805 19 3.2169 124.3317 9 0.0271 1.1894 20 3.2169 124.3317 10 0.0455 5.2998 21 3.2169 124.3317 11 0.1424 29.8293
下载: 导出CSV
表 3 灵敏度
$ {d}_{4,j} $ 随$ {R}_{j} $ 的变化Table 3. Variation of sensitivity
$ {d}_{4,j} $ with$ {R}_{j} $ ${\mathit{R} }_{15}/(\rm{N}\cdot {\rm{s} }^{2}\cdot {\rm{m} }^{-8})$ $ {\mathit{d}}_{4,15} $ ${\mathit{R} }_{18}/(\rm{N}\cdot {\rm{s} }^{2}\cdot {\rm{m} }^{-8})$ $ {\mathit{d}}_{4,18} $ ${\mathit{R} }_{5}/(\rm{N}\cdot {\rm{s} }^{2}\cdot {\rm{m} }^{-8})$ $ {\mathit{d}}_{4,5} $ ${\mathit{R} }_{8}/(\rm{N}\cdot {\rm{s} }^{2}\cdot {\rm{m} }^{-8})$ $ {\mathit{d}}_{4,8} $ 0.209 4.2291 0.296 3.5631 0.197 2.7374 0.415 1.1088 0.315 3.3323 0.425 2.7233 0.265 2.3562 0.525 0.9211 0.425 2.7041 0.525 2.2890 0.425 1.7563 0.615 0.8072 0.625 1.9785 0.625 1.9658 0.615 1.3303 0.765 0.6664 0.855 1.4831 0.875 1.4339 0.825 1.0362 0.875 0.5891 1.215 1.0379 1.125 1.1134 1.125 0.7755 1.225 0.4253 1.755 0.6906 1.875 0.6391 1.875 0.4565 1.765 0.2910 2.215 0.5254 2.125 0.5537 2.125 0.3974 2.125 0.2376 3.115 0.3454 3.145 0.3474 3.125 0.2543 3.125 0.1531 4.075 0.2448 4.125 0.2484 4.125 0.1816 4.125 0.1101
下载: 导出CSV
表 4 不同算法优化结果
Table 4. Optimization results of different algorithms
算法 风量优化平
均值/(m3∙s−1)风量优化最优
解/(m3∙s−1)平均收敛
代次数平均运行
时间/sPSO算法 13.2015 13.3148 56 16.71 标准BAS算法 13.4163 13.5067 87 9.46 改进BAS
算法13.9817 14.0185 93 10.13
下载: 导出CSV
表 5 优化调节后各分支风量分配
Table 5. Air volume distribution of each branch after optimal adjustment
m3/s 分支
编号最小
需风量调节后
风量分支
编号最小
需风量调节后
风量1 41.92 42.41 12 2.59 4.99 2 16.85 21.33 13 5.46 5.98 3 19.43 21.08 14 4.26 4.26 4 6.65 14.02 15 9.94 12.05 5 7.29 7.31 16 6.71 9.26 6 11.26 13.99 17 20.38 26.07 7 6.14 7.08 18 12.42 16.34 8 6.07 6.07 19 41.92 42.41 9 1.24 1.24 20 41.92 42.41 10 2.12 4.74 21 41.92 42.41 11 7.34 8.99
下载: 导出CSV
-
[1] ZHOU Lihong,YUAN Liming,THOMAS R,et al. Determination of velocity correction factors for real-time air velocity monitoring in underground mines[J]. International Journal of Coal Science & Technology,2017,4(4):322-332. [2] 周福宝,魏连江,夏同强,等. 矿井智能通风原理、关键技术及其初步实现[J]. 煤炭学报,2020,45(6):2225-2235. doi: 10.13225/j.cnki.jccs.zn20.0338ZHOU Fubao,WEI Lianjiang,XIA Tongqiang,et al. Principle,key technology and preliminary realization of mine intelligent ventilation[J]. Journal of China Coal Society,2020,45(6):2225-2235. doi: 10.13225/j.cnki.jccs.zn20.0338 [3] 钟德云,王李管,毕林,等. 基于回路风量法的复杂矿井通风网络解算算法[J]. 煤炭学报,2015,40(2):365-370. doi: 10.13225/j.cnki.jccs.2014.0286ZHONG Deyun,WANG Liguan,BI Lin,et al. Algorithm of complex ventilation network solution based on circuit air-quantity method[J]. Journal of China Coal Society,2015,40(2):365-370. doi: 10.13225/j.cnki.jccs.2014.0286 [4] KARACAN C O. Development and application of reservoir models and artificial neural networks for optimizing ventilation air requirements in development mining of coal seams[J]. International Journal of Coal Geology,2007,72(3/4):221-239. [5] BABU V R, MAITY T, BURMAN S. Energy saving possibilities of mine ventilation fan using particle swarm optimization[C]. IEEE International Conference on Electrical, Electronics, and Optimization Techniques （ICEEOT）, Chennai, 2016: 676-681. [6] 吴新忠,张兆龙,程健维,等. 矿井通风网络的多种群自适应粒子群算法优化研究[J]. 煤炭工程,2019,51(2):75-81.WU Xinzhong,ZHANG Zhaolong,CHENG Jianwei,et al. Optimization of multi-group self-adaptive particle swarm algorithm for mine ventilation network[J]. Coal Engineering,2019,51(2):75-81. [7] 崔传波,蒋曙光,王凯,等. 基于风量可调度的矿井风量调节[J]. 工矿自动化,2016,42(2):39-43.CUI Chuanbo,JIANG Shuguang,WANG Kai,et al. Adjustment of mine air volume based on air volume dispatchable model[J]. Industry and Mine Automation,2016,42(2):39-43. [8] 卢新明,尹红. 矿井通风智能化理论与技术[J]. 煤炭学报,2020,45(6):2236-2247. doi: 10.13225/j.cnki.jccs.ZN20.0365LU Xinming,YIN Hong. The intelligent theory and technology of mine ventilation[J]. Journal of China Coal Society,2020,45(6):2236-2247. doi: 10.13225/j.cnki.jccs.ZN20.0365 [9] 王凯,蒋曙光,马小平,等. 瓦斯爆炸致灾通风系统实验及应急救援方法[J]. 中国矿业大学学报,2015,44(4):617-622,643.WANG Kai,JIANG Shuguang,MA Xiaoping,et al. Experimental of the ventilation systems hazard by gas explosion and the methods of emergency rescue[J]. Journal of China University of Mining & Technology,2015,44(4):617-622,643. [10] 尹晓玉,谢贤平,李建功. 矿井通风网路的线性解算法及其程序设计[J]. 安全与环境学报,2015,15(4):69-73. doi: 10.13637/j.issn.1009-6094.2015.04.015YIN Xiaoyu,XIE Xianping,LI Jiangong. Linear algorithm for the mining ventilation network and program[J]. Journal of Safety and Environment,2015,15(4):69-73. doi: 10.13637/j.issn.1009-6094.2015.04.015 [11] 丰胜成,付华. 矿井通风网络风量调节分支的优化选择[J]. 辽宁工程技术大学学报(自然科学版),2019,38(6):513-516.FENG Shengcheng,FU Hua. The optimal selection of air flow adjustment branch of mine ventilation network[J]. Journal of Liaoning Technical University (Natural Science),2019,38(6):513-516. [12] ZHOU Hongbing,ZENG Xianlin,HONG Yiguang. Adaptive exact penaly design for constrained distributed optimization[J]. IEEE Transactions on Automatic Control,2019,64(11):4661-4667. doi: 10.1109/TAC.2019.2902612 [13] KHAN A H,CAO X W,LI S,et al. BAS-ADAM:an ADAM based approach to improve the performance of beetle antennae search optimizer[J]. IEEE/CAA Journal of Automatica Sinica,2020,7(2):461-471. doi: 10.1109/JAS.2020.1003048 [14] JIANG Xiangyuan, LI Shuai. BAS: beetle antennae search algorithm for optimization problems[J]. International Journal of Robotics and Control, 2018, 1（1）. DOI: 10.5430/ijrc.v1n1p1. [15] 戴英彪. 基于拉丁超立方试验设计的事故再现结果不确定性分析[D]. 长沙: 长沙理工大学, 2011.DAI Yingbiao. Uncertainty analysis of accident recurrence results based on Latin hypercube experimental design[D]. Changsha: Changsha University of Science and Technology, 2011. [16] FAN Chaojun,LI Sheng,LUO Mingkun,et al. Coal and gas outburst dynamic system[J]. International Journal of Mining Science and Technology,2017,27(1):49-55. doi: 10.1016/j.ijmst.2016.11.003 -
点击查看大图
图(7) / 表(5)
计量
- 文章访问数: 250
- HTML全文浏览量: 34
- PDF下载量: 32
- 被引次数: 0
