摘要
在地震属性中 ,波形的总体变化及其分布规律是最重要的地震参数 ,而且也是经常被人们忽视的参数。用波形建立地震相就是利用神经网络技术把地震信号的总体变化定量地刻画出来 ,简单地说就是对波形进行分类。本文简要地介绍了这一方法的基本原理 ,一般流程 ,关键参数 ,实验效果以及对过程的质量控制等关键的技术环节。实践证明地震波形包含了大量的地质信息 ,拾取并识别出这些信息是地震资料解释过程中的重要工作 ,此法有利于储层预测和油藏描述及提高钻探成功率。
In seismic attributes,the general variation of waveform and its distributed rule is most important seismic parameter,which is often neglected by the people.Using waveform to create seismic facies is to use neural network technique to quantitatively describe the general variation of seismic signal,more simply,that is to arrange the waveform under categories.The paper briefly introduced the basic principle of the method,its general flow,key parameters test effects and key technical links controlling quality of process.The practice showed that the seismic waveform includes a lot of geologic information,picking and identifying the information are important work in a process of seismic data interpretation.This method is of benefit to reservoir prediction,reservoir description and improving drilling successful rate.
出处
《石油地球物理勘探》
EI
CSCD
北大核心
2004年第5期539-542,共4页
Oil Geophysical Prospecting
