摘要
地震勘探的BSR识别技术是发现海洋天然气水合物的最经济、快捷、方便、有效的方法.在地震处理识别上,对子波的精确处理是水合物地震资料处理中最关键的一个环节,采用最小平方误差准则,即利用实际输出与期望输出的误差平方和为最小的条件,来确定反滤波因子,因此又称为最小平方子波整形.在地震处理程序中引入三种期望输出,即俞氏子波、雷克子波、Buttworth子波,对子波零相位化有较好的效果.根据上面的原理,我们开发了一套最小平方反滤波地震处理软件,对天然气水合物地震勘探资料进行试处理的结果表明,该软件在提高分辨率的同时保持了较高的信噪比.
In order to improve the resolution of seismic profiles in vertical, the aim of deconvolution is to suppress the ground filtering in exploration. The least-squares error is adopted, that is, considering sum of all anticipant output error as the least condition,to decide the inverse filtering factor. Three kinds of anticipation output is introduced: YU Shi wavelet ,Ricker wavelet, Buttworth wavelet. These wavelets help to improve the effect of zero-phase. According to the principle above, a software for least-squares-inverse filtering is developed. The results indicate this method not only improves the resolution but also maintains a relatively high ratio of signal-to-noise, especially in the detection of gas hydrate.
出处
《海洋学报》
CAS
CSCD
北大核心
2004年第3期75-81,共7页
基金
国土资源部百名人才计划资助项目.