首页|期刊导航|工矿自动化|矿井智能通风系统架构及关键技术研究

矿井智能通风系统架构及关键技术研究OA

Architecture and key technologies of intelligent mine ventilation system

中文摘要英文摘要

现有矿井智能通风研究在关键参数精准感知、多源信息耦合解算及协同调控方面仍存在不足,尚未形成一体化技术体系,难以满足矿井通风系统精细化智能管控的需求.针对上述问题,采用"感知—传输—分析—决策—应用"的一体化体系架构,设计了一种矿井智能通风系统.该系统采用基于超声波时差法的风速监测方法和基于压阻式微机电系统(MEMS)的风压监测方法,实现通风参数高精度在线感知;通过风门与风窗结构优化、动力装备变频驱动、多参数感知与智能控制技术,实现通风构筑设施远程联控与动力装备协同调节;结合多源信息融合、通风网络解算、径向基函数(RBF)神经网络识别及模糊推理等方法,进行矿井通风网络异常诊断,实现通风异常识别、位置判定及灾变影响范围判断;利用反向传播(BP)神经网络及粒子群优化(PSO)算法,结合动态权重优化机制,实现矿井需风量动态预测与通风网络全局寻优调控.现场应用结果表明,该系统风量解算最大相对误差为8.38%,风阻解算平均相对误差为2.11%,能够准确反映矿井通风网络运行状态,并可依据灾变信息自动完成应急控风并引导人员逃生,有效提升了矿井通风智能化管控水平与本质安全能力.

Existing studies on intelligent mine ventilation still show deficiencies in accurate perception of key parameters,coupled processing of multi-source information,and coordinated regulation,and have not yet formed an integrated technical system,which makes it difficult to meet the demand for refined and intelligent management of mine ventilation systems.To address these problems,an integrated system architecture of"perception-transmission-analysis-decision-application"was adopted,and an intelligent mine ventilation system was de-signed.The system adopted a wind velocity monitoring method based on the ultrasonic time difference principle and a wind pressure monitoring method based on piezoresistive Micro-Electro-Mechanical Systems(MEMS),achieving high-precision online perception of ventilation parameters.Through structural optimization of air doors and air windows,variable frequency drive of power equipment,and multi-parameter perception and intelligent control technologies,remote coordinated control of ventilation facilities and collaborative regulation of power equipment were realized.Combined with multi-source information fusion,ventilation network calculation,Radial Basis Function(RBF)neural-network-based identification,and fuzzy inference methods,abnormal diagnosis of the mine ventilation network was conducted,enabling identification of ventilation anomalies,determination of their locations,and assessment of the affected range of disasters.Backpropagation(BP)neural network and Particle Swarm Optimization(PSO)algorithms were used,together with a dynamic weight optimization mechanism,dynamic prediction of required air volume and global optimal regulation of the ventilation network were achieved.Field application results showed that the maximum relative error of air volume calculation was 8.38%,and the average relative error of wind resistance calculation was 2.11%,indicating that the system accurately reflected the operating state of the mine ventilation network,and automatically performed emergency ventilation control and guided personnel evacuation based on disaster information,effectively improving the intelligent management level and intrinsic safety capability of the mine ventilation system.

徐雪战;邹云龙

煤矿灾害防控全国重点实验室,重庆 400037||中煤科工集团重庆研究院有限公司,重庆 400039||安徽理工大学安全科学与工程学院,安徽淮南 232001煤矿灾害防控全国重点实验室,重庆 400037||中煤科工集团重庆研究院有限公司,重庆 400039

矿业与冶金

矿井智能通风通风参数监测通风网络解算百叶式自动调节风窗按需供风灾变控风

intelligent mine ventilationventilation parameter monitoringventilation network calculationlouvered automatic regulating air windowdemand-based air supplyemergency ventilation control

《工矿自动化》 2026 (2)

7-15,9

煤矿安全国家工程中心原始创新能力提升项目(2502-500106-04-02-926200)重庆市英才计划项目(CQYC20210301417)天地科技股份有限公司科技创新创业资金专项项目(2022-2-TD-ZD010).

10.13272/j.issn.1671-251x.2025110068

评论