摘要
通过GeoEast系统中地震反演技术在鄂尔多斯盆地BD三维地震工区煤层勘探中的应用,总结出通过地震反演技术识别煤层分布和厚度的流程。应用结果表明,利用GeoEast系统中的基于遗传算法的BP神经网络叠后反演预测煤层分布范围,利用叠前弹性参数反演预测纵、横波速度,进而可预测煤层厚度。反演结果与测井数据吻合度高,结合相关地震资料信息,能够精细描述煤层气空间分布形态及其厚度特征。
We use seismic inversions provided by GeoEast on 3Dseismic data in BD area,Erdos Basin for coal exploration and set up aprediction workflow for coal bed distribution and thickness.We perform post-stack BP neural network inversion based on genetic algorithm to predict coal bed distribution,and the pre-stack elastic parameter inversion to predict primary and shear wave velocity and then coal bed thickness.The prediction results coincide very well with well logs,and it can describe the coal distribution and coal bed thickness with other seismic information.
出处
《石油地球物理勘探》
EI
CSCD
北大核心
2014年第A01期184-191,8,共8页
Oil Geophysical Prospecting
基金
国家科技重大专项课题(2011ZX05019-003)资助
关键词
煤层预测
煤层厚度
BP神经网络
弹性参数
coal bed prediction,coal bed thickness,BP neural network,elastic parameter