利用岩性地震反演信息划分煤体结构  

Classification of coal body structure using seismic lithology inversion information

作　　者：彭刘亚[1,2] 崔若飞[1,2] 任川[2] 崔大尉[1,2] 

机构地区：[1]中国矿业大学煤层气资源与成藏过程教育部重点实验室,江苏徐州221008 [2]中国矿业大学资源与地球科学学院,江苏徐州221116

出　　处：《煤炭学报》2013年第A02期410-415,共6页Journal of China Coal Society

基　　金：国家自然科学基金煤炭联合基金重点资助项目(u1261202)

摘　　要：常规的煤田地震勘探技术主要解决煤层的构造问题,无法对于瓦斯突出相关的构造煤发育情况、煤岩层的岩性以及煤体结构作出评价。从岩石物理学的角度出发,通过分析对比不同煤体结构的岩性参数差异,提出了煤体结构破坏指示因子的概念。利用三维地震资料、测井曲线进行约束反演获得的波阻抗数据体作为外部属性,并结合其他地震属性,训练概率神经网络对煤体结构破坏指示因子进行反演预测,从而划分煤体结构。结果表明岩性反演结果与矿井瓦斯分布关联度较高。Abstract：Conventional coalfield seismic prospecting technology is mainly used for the interpretation of coal seam structures, it is unable to assess and evaluate the tectonic coal development, the lithology of coal seam and its surrounding rock, and coal body structure. In this paper, the concept of coal body damage indicator is proposed using lithology parameters comparison among different coal body structures on the basis of rock physics. Acoustic wave imped- ance, as an external attribute, acquired by inversion with both seismic data and well log, and other seismic attributes are utilized to train a probabilistic neural network in order to calculate and predict the coal body damage indicator, which is used for classifying coal body structures. The case study results show that the lithology inversion is highly in consistent with actual gas distribution of the coalmine, which indicates that the method is of high significance to the gas prediction and control in the eoalmine. Key words： gas outburst ; coal body structure ; lithology inversion ; indicator; probabilistie neural network

关 键 词：瓦斯突出 煤体结构 岩性反演 指示因子 概率神经网络 

分 类 号：P631.4[天文地球—地质矿产勘探]