神经网络瓦斯涌出量预测系统研究与应用

Research and application of neural network gas emission prediction system

采集煤矿瓦斯监控系统“一通三防”自动化监测系统实时数据、通风技术数据、抽采技术数据、防突技术数据、地测技术数据、产量进尺数据,并与技术数据相融合,生成“一通三防”综合服务数据宽表,建立“一通三防”数据管理系统。通过神经网络瓦斯涌出量模型对海量“一通三防”数据进行自主学习与感知数据,实现瓦斯涌出量预测,自动编写瓦斯数据分析报告。

Collect real-time data of"one link and three prevention"automatic monitoring system such as coal mine gas monitoring system,ventilation section technical data,extraction section technical data,anti-breakthrough technology data,geodesy technical data,and output advance data.Real-time data and technical data are fused to generate a wide table of"one link and three defense"comprehensive service data,and establish a"one link and three defense"data middle platform.Through the neural network gas outflow model[1-2],the massive"one link and three defense"data is independently learned and perceived to realize the prediction of gas outflow and automatically compile gas data analysis reports.