基于边缘计算的采煤工作面甲烷监测模式研究

Research on methane monitoring mode of coal mining face based on edge computing

针对工作面现有甲烷监测点不足,无法满足智慧矿山建设中工作面甲烷感知、分析和预警需求的问题,结合边缘计算技术、矿山物联网环境感知技术和数据传输等相关技术,分析了甲烷微功耗感知、自组网传输、监测终端供电和传感位置服务问题。提出了集微功耗甲烷感知、自供电、自定位及自组网数据传输于一体的甲烷监测终端,配合工作面区域边缘网关、矿端边缘服务器和云端大数据分析,建立了基于云、边、端三级的甲烷边缘监测模式的系统架构,解决了工作面区域基于边缘计算的甲烷感知和传输问题。以100台液压支架、2台端头支架工作面为例,在笛卡尔二维坐标系下设计了该工作面基于边缘计算的甲烷监测方案,实现了甲烷浓度二维空间范围的大数据量监测,提出了工作面区域全覆盖的甲烷含量数字场实现方法和云图直观展示方式。在甲烷含量空间场和云图展示的基础上,提出了2种基于大数据支撑的甲烷含量分析预警模式和思路,一是对甲烷超限含量前后区域空间场进行关联分析,提取工作面甲烷含量超限对应空间场的分布特征和变化规律,利用区域含量场及早发现甲烷含量超限问题,另一种是利用大数据分析工具解决甲烷监测报警阈值以下的显现异常问题。

In view of the shortage of existing methane monitoring points in the coal face,and it can notmeet the the needs of methane sensing,analysis and early warning in the coal face of smart mine construction.In combination with the edge of computing technology,mine Internet of things environment sensing technology and data transmission related technologies,analysis methane micropower perception,network transmission,monitoring terminal power supply and sensing location service issues,a methane monitoring terminal that integratesmicro-power methane perception,self-powered supply,self-positioning,and self-organizing network data transmission was proposed.With the analysis of the work area regional edge gateway,mine-end edge servers and cloud big data,a system architecture of methane edge monitoring mode was established based on cloud-edge and end to solve the problem of methane sensing and transmission based on edge computing in the face area.Taking the 100 hydraulic supports and 2 end support of coal faces as an example,a methane monitoring scheme based on edge calculation was designed in the Cartesian two-dimensional coordinate system to realize the monitoring of large amounts of data in the two-dimensional spatial range of methane concentration.The realization method of the methane concentration digital field with full coverage in the coal face area and the visual display method of the cloud image were proposed.On the basis of the spatial field of methane concentration and the cloud image display,two early warning modes and ideas for methane analysis based on big data support were proposed.One is to perform correlation analysis of regional space field before and after methane over-limit concentration,and extract the corresponding methane over-limit in the coal face.The distribution characteristics and changing rules of the spatial field use the regional concentration field to detect the methane concentration over-limit problem as early as possible.Another one is to use a big data analysis tool to solve the problem of abnormal manifestation below the methane monitoring alarm threshold.