中文摘要：

为有效分析矿井瓦斯监测数据以拓展监控系统功能,实现工作面瓦斯浓度的有效预警,研究基于多测点实时监测数据关联分析的工作面瓦斯（甲烷）浓度预警方法。通过分析瓦斯实测数据的统计特征,以及利用贝叶斯网络方法分析工作面与其关联监测点瓦斯实测数据构成时间序列的关联特征,确定基于实时监测数据的瓦斯预警指标及其预警阈值,进而分析瓦斯浓度异常情况,实现基于监测数据分析的实时、动态量化预警。实例分析表明,将该方法应用于工作面瓦斯浓度预警,结果显示了瓦斯浓度持续偏大时段反映出的异常情况,符合实际瓦斯浓度变化趋势。

英文摘要：

For effective analysis of coal mine gas monitoring data to realize effective working face gas concentration pre-warning and expand the monitoring system functions,a monitoring data correlation analysis-based working face gas concentration pre-warning method was worked out. The correlation characteristics of multi- points monitoring data at working face were abstracted by using Baysian networks method,combining its statistical characteristics. Real-time monitoring data-based pre-warning indexes and their thresholds were determined and the gas concentration abnormal situations could be analyzed to realize timely and dynamically quantitative pre-warning. The pre-warning method was used to carry out gas concentration pre-warning analysis at the working face 32212 of a certain coal mine in Ningxia. The analysis results conform with the reality.