基于吸引−排斥算法的精准分风方案决策

倪景峰; 陈敦炜; 刘玉娇

doi:10.13272/j.issn.1671-251x.2024110006

基于吸引−排斥算法的精准分风方案决策

倪景峰^{1, 2,},
陈敦炜^{1, 2},
刘玉娇^{1, 2}

1.
辽宁工程技术大学安全科学与工程学院，辽宁阜新　123000
2.
辽宁工程技术大学矿山热动力灾害与防治教育部重点实验室，辽宁阜新　123000

基金项目:

国家自然科学基金青年基金项目(52104194)。

详细信息

作者简介:
倪景峰（1975—），男，河北承德人，教授，博士，主要从事通风（流体）网络算法、矿井智能通风、矿井灾变通风与救援等方面的研究工作，E-mail：njf_3829@163.com

中图分类号: TD724
计量
- 文章访问数: 29
- HTML全文浏览量: 4
- PDF下载量: 5
出版历程
- 收稿日期: 2024-11-03
- 修回日期: 2025-04-13
- 网络出版日期: 2025-04-08
- 刊出日期: 2025-04-14

Precise wind allocation scheme decision based on attraction-repulsion algorithm

NI Jingfeng^{1, 2,},
CHEN Dunwei^{1, 2},
LIU Yujiao^{1, 2}

1.
Safety Science and Engineering College, Liaoning Technical University, Fuxin 123000, China
2.
Key Laboratory of Mine Thermodynamic Disasters and Control of Ministry of Education, Liaoning Technical University, Fuxin 123000, China

摘要

摘要:
为解决井下生产作业过程中由于矿井通风设施和风网结构变化导致的通风系统分支风量波动，进而引发的用风地点风量不足问题，提出了一种基于吸引−排斥算法（AROA）的精准分风方法。以通风机功耗最小化为优化目标，工作面与备用面的需风量为约束条件，建立矿井通风系统数学模型。采用AROA，通过精准调控通风机及井下既有通风设施，迭代生成优化解。优化过程中，融合改进布朗运动、三角函数变换、随机解选择机制与记忆型局部搜索算子，对候选解实施动态筛选与精准调优，最终实现通风运行成本最优的精准分风方案。性能测试结果表明：与遗传算法（GA）、模拟退火−改进粒子群算法（SA−IPSO）和单调盆地跳跃算法（MBH）相比，AROA在综合寻优性能方面优势显著；在求解Ackley函数时，其获取最优解与平均最优解所经历的迭代次数均少于GA，SA−IPSO和MBH。实例分析结果表明：采用基于AROA的精准分风算法所确定的精准分风方案后，风窗面积调节量达50.4%；左翼通风机功率从131.72 kW降至97.95 kW，降幅达25.6%；右翼通风机功率从188.22 kW降至146.62 kW，降幅达22.1%；总节能率达23.56%。某煤矿实际应用结果表明：采用基于AROA的精准分风算法所确定的精准分风方案后，通风机风量降低了11.2%，通风机风压下降了10.1%，功率降低了20.7%。
- 矿井通风 /
- 精准分风 /
- 吸引−排斥算法 /
- 分窗面积 /
- 通风机风压 /
- 通风机分量
Abstract:
To address the issue of fluctuating branch airflow in the ventilation system caused by changes in mine ventilation facilities and air network structure during underground production operations, which in turn leads to insufficient airflow at consumption points, a precise wind allocation algorithm based on the Attraction-Repulsion Optimization Algorithm (AROA) is proposed. The ventilation fan power consumption minimization was set as the optimization objective, with the required airflow for working and standby faces as constraints, and a mathematical model of the mine ventilation system was established. By employing AROA, the ventilation fan and existing underground ventilation facilities were precisely controlled, and an optimized solution was iteratively generated. During the optimization process, an improved Brownian motion, trigonometric function transformation, random solution selection mechanism, and memory-based local search operator were integrated to dynamically filter and fine-tune candidate solutions, ultimately achieving an optimal precise wind allocation plan with the lowest ventilation operation cost. Performance test results showed that AROA had a significant advantage in comprehensive optimization performance compared to Genetic Algorithm (GA), Simulated Annealing-Improved Particle Swarm Optimization (SA-IPSO), and Monotonic Basin Hopping (MBH). When solving the Ackley function, AROA required fewer iterations to obtain the optimal and average optimal solutions compared to GA, SA-IPSO, and MBH. Case study results showed that the precise wind allocation scheme determined by the AROA-based algorithm resulted in a 50.4% adjustment in the air window area. The left-wing fan power decreased from 131.72 kW to 97.95 kW (a reduction of 25.6%), and the right-wing fan power decreased from 188.22 kW to 146.62 kW (a reduction of 22.1%), achieving a total energy-saving rate of 23.56%. Actual application results in a coal mine demonstrated that the AROA-based algorithm reduced the fan airflow by 11.2%, while the fan air pressure decreased by 10.1%, ultimately achieving a 20.7% reduction in power consumption. The precise wind allocation scheme determined by the AROA-based algorithm reduced fan air pressure by 10.1%, fan airflow by 11.2%, and power by 20.7%.
- mine ventilation /
- precise wind allocation /
- Attraction-Repulsion Algorithm /
- window area /
- fan air pressure /
- fan airflow

HTML全文

图 1 候选解的相互作用及其位置更新机制

Figure 1. Interaction of candidate solutions and their position update mechanisms

下载: 全尺寸图片幻灯片

图 2 平均最优解曲线

Figure 2. Average optimal solution curves

下载: 全尺寸图片幻灯片

图 3 算法收敛曲线

Figure 3. Algorithm convergence curves

下载: 全尺寸图片幻灯片

图 4 算法执行时间

Figure 4. Algorithm execution time

下载: 全尺寸图片幻灯片

图 5 最优解分布

Figure 5. Optimal solution distribution

下载: 全尺寸图片幻灯片

图 6 矿井通风网络拓扑结构

Figure 6. Mine ventilation network topology

下载: 全尺寸图片幻灯片

图 7 左翼通风机功率曲线

Figure 7. Left wing ventilator power curves

下载: 全尺寸图片幻灯片

图 8 右翼通风机功率曲线

Figure 8. Right wing ventilator power curves

下载: 全尺寸图片幻灯片

图 9 唐安煤矿通风网络拓扑结构

Figure 9. Topology of the ventilation network of Tang'an Coal Mine

下载: 全尺寸图片幻灯片

表 1 矿井通风网络基本参数

Table 1 Basic parameters of mine ventilation network

分支编号	始节点	末节点	风阻/ （kg·m⁻⁷）	风量/ （m³∙s⁻¹）	风窗面积/m²	是否可调节
e₁	5	6	0.13	16	1.37	是
e₅	16	17	0.11	100	—	是
e₆	7	8	0.13	80	—	是
e₁₄	9	16	0.14	25	1.5	是
e₁₅	13	14	0.08	8	0.68	是
e₁₆	12	15	0.09	3	0.23	是
e₂₂	4	6	0.02	19.28	1.45	是
e₂₃	3	7	1.81	12.72	1.03	是

下载: 导出CSV

表 2 巷道精准分风方案结果统计

Table 2 Statistics of precision wind allocation scheme for roadways

分支编号	GA算法			SA−IPSO算法			MBH算法			AROA算法
分支编号	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²
e₁	16.00	169	1.49	16.00	169	1.49	16	169	1.39	16	169	1.49
e₁₄	20.00	349	1.30	12.00	385	0.74	8	396	0.49	6	400	0.36
e₁₅	8.00	198	0.69	8.00	169	1.49	8	198	0.69	8	198	0.69
e₁₆	3.00	296	0.21	3.00	296	0.21	3	296	0.21	3	296	0.21
e₂₂	11.48	209	0.96	4.41	211	0.37	2	211	0.17	2	211	0.17

下载: 导出CSV

表 3 各算法运行效果比较

Table 3 Comparison of algorithm performance

算法	左翼通风			右翼通风机			节能率/%
算法	功率/kW	风量/（m³·s⁻¹）	风压/Pa	功率/kW	风量/（m³·s⁻¹）	风压/Pa	节能率/%
GA	131.72	75	1405	188.22	95	1585	—
SA−IPSO	114.07	67	1362	167.21	87	1532	13.05
MBH	103.70	62	1338	154.21	83	1490	5.17
AROA	97.95	60	1306	146.62	81	1457	23.56

下载: 导出CSV

表 4 唐安煤矿通风网络调节参数

Table 4 Adjustment parameters for the ventilation network of Tang'an Coal Mine

分支编号	始节点	末节点	风阻/ （kg·m⁻⁷）	风量/ （m³∙s⁻¹）	风窗面积/m²	是否可调节
e₇	6	7	0.75	19.88	3.05	是
e₈	7	8	0.04	3.76	0.15	是
e₂₂	25	50	0.01	1.77	0.06	是
e₃₀	59	14	0.64	21.88	0.86	是
e₄₀	24	21	0.30	10.40	1.53	是
e₄₆	31	21	0.01	0.20	0.03	是
e₄₉	32	10	0.08	5.91	0.37	是
e₅₁	33	50	0.20	5.09	0.15	是
e₅₉	40	41	0.04	3.34	0.22	是
e₆₂	35	41	0.01	0.27	0.05	是
e₆₆	44	45	0.58	21.16	1.80	是
e₆₈	30	81	0.01	0.20	0.02	是
e₆₉	81	13	0.28	14.00	0.52	是
e₇₀	29	59	0.01	0.20	0.01	是
e₈₃	47	48	0.01	1.69	1.36	是
e₈₅	77	78	0.01	1.71	0.23	是
e₈₆	78	74	0.06	5.01	0.18	是
e₈₈	79	80	0.04	3.68	0.13	是
e₉₅	75	73	0.01	0.20	0.04	是
e₉₇	61	66	0.94	8.37	1.17	是
e₁₀₈	66	65	0.03	2.20	0.41	是
e₁₁₂	52	51	0.28	8.71	2.33	是
e₁₁₅	65	72	2.78	18.57	0.60	是
e₁₁₇	69	70	0.06	4.31	0.15	是
e₁₁₈	68	71	0.25	6.68	0.23	是
e₁₂₁	72	73	2.34	15.40	0.54	是
e₁₂₄	34	42	0.01	0.20	0.01	是
e₁₂₇	85	—	30.83	188.35	—	是

下载: 导出CSV

表 5 唐安煤矿精准分风方案结果统计

Table 5 Statistics of precision wind allocation scheme for Tang’an Coal Mine

分支编号	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²	分支编号	风量/ （m³∙s⁻¹）	调节阻力/Pa	风窗面积/m²
e₇	2.85	302	0.19	e₈₃	2.27	4	1.22
e₈	4.72	699	0.21	e₈₅	13.42	26	2.08
e₂₂	0.62	1367	0.02	e₈₆	14.58	1076	0.52
e₃₀	18.03	928	0.68	e₈₈	2.40	1111	0.09
e₄₀	8.32	41	1.46	e₉₅	3.75	289	0.26
e₄₆	2.7	69	0.38	e₉₇	3.69	1513	0.11
e₄₉	2.34	507	0.12	e₁₀₈	1.56	34	0.31
e₅₁	2.95	1332	0.10	e₁₁₂	3.23	15	0.95
e₅₉	2.70	459	0.15	e₁₁₅	16.56	1353	0.51
e₆₂	3.75	444	0.21	e₁₁₇	3.51	1362	0.11
e₆₆	—	—	—	e₁₁₈	1.74	1360	0.06
e₆₈	2.22	192	0.19	e₁₂₁	7.30	1073	0.26
e₆₉	15.70	944	0.59	e₁₂₄	2.70	992	0.11
e₇₀	2.25	208	0.18	e₁₂₇	167.30	—	—

下载: 导出CSV

参考文献(21)

[1]	陈林,谢勇强,罗云勇. 矿井通风网络优化的调节与实现[J]. 昆明冶金高等专科学校学报,2022,38(6):30-35. CHEN Lin,XIE Yongqiang,LUO Yunyong. Research and realization of the optimization of mine ventilation network[J]. Journal of Kunming Metallurgy College,2022,38(6):30-35.
[2]	司俊鸿,王小军,黄欣. 基于改进DE算法的矿井通风网络非线性优化求解[J]. 煤炭工程,2016,48(3):64-67. SI Junhong,WANG Xiaojun,HUANG Xin. A modified differential evolution algorithm for nonlinear optimization solution of mine ventilation network[J]. Coal Engineering,2016,48(3):64-67.
[3]	吴新忠,张芝超,王凯,等. 基于DE−GWO算法的矿井风网风量调节方法[J]. 中南大学学报(自然科学版),2021,52(11):3981-3989. DOI: 10.11817/j.issn.1672-7207.2021.11.019 WU Xinzhong,ZHANG Zhichao,WANG Kai,et al. Method for adjusting air volume of mine ventilation network based on DE-GWO algorithm[J]. Journal of Central South University (Science and Technology),2021,52(11):3981-3989. DOI: 10.11817/j.issn.1672-7207.2021.11.019
[4]	王海宁,彭斌,彭家兰,等. 基于三维仿真的矿井通风系统及其优化研究[J]. 中国安全科学学报,2013,23(9):123-128. DOI: 10.3969/j.issn.1003-3033.2013.09.021 WANG Haining,PENG Bin,PENG Jialan,et al. Study on mine ventilation system and its optimization based on 3D simulation[J]. China Safety Science Journal,2013,23(9):123-128. DOI: 10.3969/j.issn.1003-3033.2013.09.021
[5]	吴新忠,胡建豪,魏连江,等. 矿井通风网络的反向增强型烟花算法优化研究[J]. 工矿自动化,2019,45(10):17-22,67. WU Xinzhong,HU Jianhao,WEI Lianjiang,et al. Research on opposition-based enhanced fireworks algorithm optimization for mine ventilation network[J]. Industry and Mine Automation,2019,45(10):17-22,67.
[6]	宋佳林. IPSO−TS算法在矿井通风网络风量优化中的应用研究[J]. 矿业安全与环保,2022,49(2):78-82. SONG Jialin. Application of IPSO-TS algorithm in air volume optimization of mine ventilation network[J]. Mining Safety & Environmental Protection,2022,49(2):78-82.
[7]	张兴国,周玉. 基于ACPSO算法的矿井通风网络解算研究[J]. 辽宁工程技术大学学报(社会科学版),2018,20(4):305-311. DOI: 10.11955/j.issn.1008-391x.20180410 ZHANG Xingguo,ZHOU Yu. Study on ACPSO algorithm for mine ventilation network[J]. Journal of Liaoning Technical University(Social Science Edition),2018,20(4):305-311. DOI: 10.11955/j.issn.1008-391x.20180410
[8]	郭一楠,王春,杨继超. 基于文化粒子群优化算法的矿井通风网络[J]. 东南大学学报(自然科学版),2013,43(增刊1):48-53. GUO Yinan,WANG Chun,YANG Jichao. Mine ventilation network based on cultural particle swarm optimization algorithm[J]. Journal of Southeast University (Natural Science Edition),2013,43(S1):48-53.
[9]	LI Junqiao,LI Yucheng,ZHANG Wei,et al. Multi- objective intelligent decision and linkage control algorithm for mine ventilation[J]. Energies,2022,15(21). DOI: 10.3390/EN15217980.
[10]	韩正化. 基于麻雀搜索算法的矿井风量智能优化调控[D]. 徐州:中国矿业大学,2022. HAN Zhenghua. Intelligent optimal regulation of mine air volume based on sparrow search algorithm[D]. Xuzhou:China University of Mining and Technology,2022.
[11]	厍向阳,史经俭,罗晓霞. 栅格数据模型中附有条件的最短路径算法[J]. 计算机应用,2008(4):856-859. SHE Xiangyang,SHI Jingjian,LUO Xiaoxia. Shortest path algorithm confined to conditions in grid data model[J]. Journal of Computer Applications,2008(4):856-859.
[12]	WANG Jinmiao,XIAO Jun,XUE Yan,et al. Optimization of airflow distribution in mine ventilation networks using the modified sooty tern optimization algorithm[J]. Mining,Metallurgy & Exploration,2023,41(1):239-257.
[13]	CYMERYS K,OSZUST M. Attraction–repulsion optimization algorithm for global optimization problems[J]. Swarm and Evolutionary Computation,2024. DOI: 10.1016/J.SWEVO.2023.101459.
[14]	KENNEDY J,EBERHART R. Particle swarm optimization[C]. International Conference on Neural Networks,Perth,1995:1942-1948.
[15]	ASKARI Q,YOUNAS I,SAEED M. Political optimizer:a novel socio-inspired meta-heuristic for global optimization[J]. Knowledge-Based Systems,2020. DOI: 10.1016/j.knosys.2020.105709.
[16]	陈乐天,袁红,孙昌璞. 布朗运动理论及其在复杂气候系统研究中的应用[J]. 物理,2022,51(9):588-601. DOI: 10.7693/wl20220901 CHEN Letian,YUAN Hong,SUN Changpu. Brownian motion theory and its application in the study of complex climate systems[J]. Physics,2022,51(9):588-601. DOI: 10.7693/wl20220901
[17]	UR R H,REHAN A,ABDUL-MAJID W,et al. Analysis of Brownian motion in stochastic schrödinger wave equation using sardar sub-equation method[J]. Optik,2023,289. DOI: 10.1016/J.IJLEO.2023.171305.
[18]	曹亚丽,余牧舟,杨俊峰,等. 一种改进的人工蜂群算法研究[J]. 现代电子技术,2020,43(12):133-137,141. CAO Yali,YU Muzhou,YANG Junfeng,et al. Research on an improved artificial bee colony algorithm[J]. Modern Electronics Technique,2020,43(12):133-137,141.
[19]	陈永强,张显涛. 基于NSGA−Ⅱ算法与离散模块梁单元水弹性方法的连接件优化设计分析[J/OL]. 中国舰船研究:1-10 [2024-10-25]. https://doi.org/10.19693/j.issn.1673-3185.04337. CHEN Yongqiang,ZHANG Xiantao. Optimization design of connectors using the discrete-module-beam hydroelasticity method and NSGA-II algorithm[J]. Chinese Journal of Ship Research:1-10 [2024-10-25]. https://doi.org/10.19693/j.issn.1673-3185.04337.
[20]	邵良杉,王振,李昌明. 基于模拟退火与改进粒子群的矿井通风优化算法[J]. 系统仿真学报,2021,33(9):2085-2094. SHAO Liangshan,WANG Zhen,LI Changming. Optimization algorithm of mine ventilation based on SA-IPSO[J]. Journal of System Simulation,2021,33(9):2085-2094.
[21]	孟思彤. 基于单调盆地跳跃算法的矿井风量优化研究[D]. 阜新:辽宁工程技术大学,2023. MENG Sitong. Study on optimization of mine air volume based on monotone basin jump algorithm[D]. Fuxin:Liaoning Technical University,2023.

施引文献(20)

期刊类型引用(8)

1.	陈梓华，马占元，李敬兆. 基于RNN的煤矿安全隐患信息关键语义智能提取系统. 煤炭工程. 2021(03): 185-189 . 百度学术
2.	何桥，许金. 基于信息熵—集对分析的煤矿隐患排查治理风险评估模型. 能源与环保. 2020(01): 56-59 . 百度学术
3.	何桥，许金，陈清. 煤矿安全隐患风险评价研究. 工矿自动化. 2019(08): 48-53+59 . 本站查看
4.	张俭让，黄玉鑫，闫振国，张磊，霍小泉. 煤矿安全生产标准化信息管理系统设计. 工矿自动化. 2019(12): 81-85 . 本站查看
5.	郝丹丹，郭景峰，王燕君. 基于k-shell的社区发现算法研究. 河北省科学院学报. 2018(02): 18-29 . 百度学术
6.	吴飞. 煤矿企业隐患排查治理信息管理系统的研究与设计. 中国煤炭. 2018(05): 74-77 . 百度学术
7.	张成才. 关于煤矿重大隐患特性及治理体系的分析. 内蒙古煤炭经济. 2018(21): 102+138 . 百度学术
8.	李斌，郑淇，华圣元，徐蓉，李欣，赵克勤. 基于四元联系数的中医四诊与银屑病疗效关系研究. 辽宁中医杂志. 2018(11): 2241-2246 . 百度学术