全面感知和实时互联是智能化煤矿最基本的功能要素。现阶段采煤工作面整体环境感知能力不足，感知设备设置监测点数量有限，末端无线网络不够健全，缺乏高精度位置服务，导致矿井与采煤工作面全面感知所需数据样本量偏少，信息透明度不够，隐患识别和安全预警准确性偏低。针对该问题，以采煤工作面为应用场景，以CH₄为监测对象，研究煤矿工作环境参数大样本数据感知关键技术及监测模式。通过研究无线低功耗CH₄传感与自标校技术，实现在采煤工作面布置大量CH₄传感器进行全面感知，解决长时间免标校维护的技术难题；通过研究传感设备对象编码与定位技术，解决大量传感设备的身份和位置识别难题；通过研究适用于矿井线性空间的高速无线数据传输技术，以及无线骨干网链路节点的路由自发现、网络故障自主发现、故障节点及时隔离和自恢复技术，解决采煤工作面布设大量CH₄传感器及工作面移动带来的数据实时传输与维护问题；通过研究基于边缘计算的大样本数据连续监测模式，针对采集的大量CH₄传感数据，利用空间数字云图技术，实现整个采煤工作面CH₄面域连续监测和全面感知及作业现场数据分级处理。采煤工作面CH₄大样本数据感知关键技术及监测模式为其他矿井环境参数的全面感知研究提供了基础技术积累。

Comprehensive perception and real-time connectivity are fundamental functional elements of intelligent coal mines. Currently, coal mining faces suffer from insufficient overall environmental perception capabilities. Limitations include the small number of monitoring points for perception devices, inadequate terminal wireless network coverage, and a lack of high-precision positioning services. These shortcomings result in inadequate data sample sizes required for comprehensive perception of mines and coal mining faces, low information transparency, and reduced accuracy in hazard identification and safety warnings. To address these issues, this study investigated coal mining faces as the application scenario and CH₄ as the monitoring target, exploring key technologies and monitoring models for large-scale data perception of coal mine environmental parameters. By investigating low-power wireless CH₄ sensing and self-calibration technologies, the study enabled the deployment of numerous CH₄ sensors in coal mining faces for comprehensive perception, resolving technical challenges associated with calibration-free maintenance. The study also addressed the difficulties of identifying the identities and locations of numerous sensors by developing device encoding and positioning technologies for sensing devices. Additionally, the study proposed high-speed wireless data transmission technologies suitable for the linear space of mines, along with autonomous routing discovery, network fault detection, timely isolation of fault nodes, and self-recovery for wireless backbone link nodes. These advancements solved the real-time data transmission and maintenance challenges arising from the deployment of large numbers of CH₄ sensors and the mobility of coal mining faces. Furthermore, a continuous monitoring model for large-scale data based on edge computing was developed. This model processed the collected CH₄ sensor data using spatial digital cloud mapping technology to achieve continuous monitoring and comprehensive perception of CH₄ across the entire coal mining face, as well as hierarchical data processing at operational sites. The key technologies and monitoring model for large-scale data perception of CH₄ in coal mining faces accumulate foundational technical knowledge for comprehensive perception studies of other mine environmental parameters.