多信息约束下的二维地震相分析

2D Seismic Facies Analysis Constrained by Multi-source Information

针对二维地震资料的特点及其在地震相分析中存在的问题,提出了多信息约束下的模糊神经网络地震波形聚类方法,通过对模糊神经网络聚类算法目标函数的配置,把地震波形信息、地震资料的空间变化特征、井点沉积相类型和井点砂岩厚度等信息融合到地震波形的聚类过程中,实现了波形聚类过程的地质信息督导。通过在苏里格气田南部二维地震区的实际应用证明该方法具有较强的信息融合能力,得到的地震相成果具有较好的地质含义。

In consideration of the characteristics of seismic data and the problem of seismic facies analysis,a method based on fuzzy neural network was proposed which was constrained by multi-source information.By arranging the object function of its cluster algorithm constrained by multi-source information,the characteristics of seismic wave shape,the special changes of seismic data,the type of well sedimentary faceis,variety of seismic attribute and sand thickness in wells are integrated into the seismic wave cluster analysis,by which the geological information supervision was implemented in waveform cluster process.The application in southern Sulige Gasfield shows that the method is of high information possess ability,efficient integration and provides more geological information than the traditional method.