混采数据分离中插值与去噪的同步处理

Synchronous interpolation and denoising in simultaneous-source data separation

近年来,由于新兴的混合采集观测系统在很大程度上提高了采集效率,因此得到了很多学者和石油公司的青睐.但在实际应用中,这种特殊观测系统的采集质量却受到很多因素的影响.一方面,该观测系统采集到的炮记录会受到相邻炮记录的干扰;另一方面,复杂的采集环境使得地震记录中包含部分空道;另外,采集过程中的场地环境干扰会不可避免地带入随机噪音,它们都会影响采集质量.虽然已有一些学者对这些因素做了相关研究,但都是单独分析,未能综合考虑各种干扰因素.本文基于稀疏约束反演的基本原理,将混合炮数据的分离、缺失道集的插值以及对随机噪音的压制问题整合在一起,通过一步处理同时减小如上三方面因素的不利影响,在改善信噪比的同时极大地提高了地震资料的处理效率.文章利用模拟数据和实际数据对这种新方法进行了验证,均得到了比较满意的效果.

Much concern has been given to the simultaneous shooting because this acquisition design can improve acquisition efficiency largely. However, there are many factors affecting the acquisition quality in seismic exploring field. First of all, the seismic records may include some adjacent shot gathers destroying the seismic records. Secondly, some blank traces will be contained in records because of the complex acquisition environment; last, the disturbance around the survey field will introduce random noise which will contaminate the seismic record. The three cases mentioned above have been studied by many researchers separately before. On the basis of sparse constrained inversion we will accomplish these three cases simultaneously. To a large extent, it will increase the efficiency of the seismic data processing and improve the SNR. In this paper, we apply the simulated dataset and real dataset respectively to verify our method and have achieved promising results.