The gas disaster monitoring data is a basis for monitoring and prediction of Coal mine methane gas.This paper uses modern information processing technology and artificial intelligence platform for each multi-sensor monitoring data, including downhole temperature, ventilation rate, gas concentration, excavation, analysis, processing, synthesis, hazard signs gas extracted information, and that these symptoms with characteristics of the information matrix.In the use of maximum entropy method based on feature extraction, multi-source data fusion for multi-platform, proposed integration of rough set and neural network algorithm, achieved CMM data forecasting methods to increase the confidence of mine condition monitoring to improve the accuracy of environmental monitoring, has practical significance.%瓦斯灾害监测数据是煤矿瓦斯监控和预测的基础;采用现代信息处理技术和人工智能理论对每个监测平台的多传感器检测数据,主要包括井下温度、通风景、瓦斯浓度等进行挖掘、分析、处理、综合,提取出瓦斯灾害征兆信息,并用特征矩阵表示这些征兆信息;在利用最大熵方法进行特征提取的基础上,对多源数据进行多平台信息融合,提出粗糙集和神经网络融合算法,实现了煤矿瓦斯数据预报方法,增加了矿井状态监测置信度,提高了环境监测的准确率,具有实际应用意义.
展开▼
