基于ESP32与GD32的矿场环境监测及预警系统
An ESP32 and GD32-Based Environment Monitoring and Early Warning System for Mining Areas

针对矿场环境危险、监测复杂且无法实时预警这一问题，设计了基于ESP32的矿场环境监测与预警系统。本系统通过多种传感器采集各个参数数据，再通过搭载了RT-Thread操作系统的GD32辅助处理器传送至ESP32主控制器，并通过MQTT协议上传至OneNET云平台。系统实现了LED屏的现场显示、微信小程序远程监控、数据参数超过限定值后自动灯光报警及设备联动，并且集成了豆包AI大模型可以进行语音交互与智能问答。实验结果表明，LED屏连续72小时运行稳定；语音指令识别准确率超过88%，响应时间低于2.5秒；远程设备控制成功率不低于90%，响应延迟在1.5秒以内。本系统采用了国产芯片并搭配了国产操作系统的策略方案，为推进矿山数字化建设提供了一套自主可控的国产化解决方案。
To address the challenges of hazardous working conditions, complex monitoring requirements, and the lack of real-time early warning in mining environments, this study designs an ESP32-based mine environmental monitoring and early warning system. The system first collects multi-parameter data via a suite of sensors. These data are then transmitted to the ESP32 main controller through a GD32 auxiliary processor running the RT-Thread operating system, before being uploaded to the OneNET cloud platform using the MQTT protocol. Key functionalities of the system include on-site data visualization via LED displays, remote monitoring through a WeChat mini-program, automatic light-based alarms and equipment linkage triggered by parameter deviations beyond threshold values, and integration with the DouBao AI large language model to enable voice interaction and intelligent Q & A. Experimental results demonstrate that the LED display maintains stable operation for 72 consecutive hours; the voice command recognition accuracy exceeds 88% with a response time of less than 2.5 seconds; the success rate of remote device control is no less than 90% with a response delay within 1.5 seconds. By employing domestic chips and a domestic operating system, this system provides a self-controllable localized solution to advance the digital transformation of mining operations.