基于机器学习的煤矿安全自动化预警系统设计

Design of Coal Mine Safety Automation Early Warning System Based on Machine Learning

设计了一套基于机器学习的煤矿安全自动化预警系统,通过实时分析传感器采集的数据,训练深度学习模型,以预测安全风险。某煤矿的实验结果表明,该系统在正常生产、检修期及模拟紧急情况3种工况下均表现出良好的预警性能,准确率和召回率均在80%以上,能有效预警安全隐患。该系统为煤矿智能化安全管控提供了可行方案。

A machine learning based coal mine safety automation early warning system is designed to train a deep learning model by analyzing the data collected from sensors in real time in order to predict safety risks.The experimental results in a coal mine show that the system exhibits good early warning performance under three working conditions,namely,normal production,maintenance period and simulated emergency,with accuracy and recall above 80%,and is able to effectively warn of potential safety risks.The system provides a feasible solution for intelligent safety control in coal mines.