石油与天然气地质 ›› 2016, Vol. 37 ›› Issue (4): 591-597.doi: 10.11743/ogg20160417

• 技术方法 • 上一篇    下一篇

高阶统计约束稀疏盲反褶积技术研究与应用

张华1,2, 庞燕青4, 贺振华1, 杨波5, 何光明2,3, 巫芙蓉2, 陈爱萍2,3, 谢小翠2,3   

  1. 1. 成都理工大学 油气藏地质及开发工程国家重点实验室, 四川 成都 610059;
    2. 中国石油 川庆钻探工程有限公司 地球物理勘探公司, 四川 成都 610213;
    3. 中国石油 物探技术试验基地 山地地震技术试验基地, 四川 成都 610213;
    4. 中国石油 新疆油田分公司, 新疆 克拉玛依 834000;
    5. 中国石油 新疆油田分公司 陆梁油田作业区, 新疆 克拉玛依 834000
  • 收稿日期:2015-03-02 修回日期:2016-07-10 出版日期:2016-08-08 发布日期:2016-09-06
  • 作者简介:张华(1981-),男,博士研究生、工程师,地球物理资料处理和反演解释。E-mail:zhanghua_sc@cnpc.com.cn。
  • 基金资助:
    中国石油天然气集团公司科学研究与技术开发项目(2013E-38-08)。

Study and application of high-order statistics constrained sparse blind deconvolution technique

Zhang Hua1,2, Pang Yanqing4, He Zhenhua1, Yang Bo5, He Guangming2,3, Wu Furong2, Chen Aiping2,3, Xie Xiaocui2,3   

  1. 1. State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Chengdu University of Technology, Chengdu, Sichuan 610059, China;
    2. Geophysical Exploration Company, Chuanqing Drilling Engineering Co. Ltd., CNPC, Chengdu, Sichuan 610213, China;
    3. Mountain Geophysical Technology Test Center, CNPC, Chengdu, Sichuan 610213, China;
    4. Xinjiang Oilfield Company, CNPC, Karamay, Xinjiang 834000, China;
    5. Luliang Field Operation District, Xinjiang Oilfieled Company, Karamay, Xinjiang 834000, China
  • Received:2015-03-02 Revised:2016-07-10 Online:2016-08-08 Published:2016-09-06

摘要: 以贝叶斯基本理论为基础,利用高阶双谱域估算的高精度地震子波为约束,在修正柯西判别准则下,实现地震数据的稀疏盲反褶积拓频处理,获取比常规反褶积提频方法分辨率更高的可靠数据。模拟数值实验和多波实际数据处理结果表明:双谱域估算的子波精度较高,以该子波为先验约束的稀疏盲反褶积能极大提高纵波和转换波数据的分辨率,满足精细解释能分辨更小薄层的要求,可作为多波高分辨率处理的有效手段。

关键词: 子波估计, 提频, 反褶积, 反射系数, 地震勘探

Abstract: Based on the basic theory of Bayesian and using the high precision wavelet estimated in high-order double spectrum domain as constraints,this paper realized sparse blind deconvolution frequency-expanding processing under the modified Cauchy criteria and obtained data that are more reliable and have higher frequency than the conventional deconvolution frequency-enhancing method.The results of numerical experiments and real data processing show that the wavelet estimated in the dual spectrum domain has higher accuracy,under the priori constraint of which the sparse blind deconvolution can greatly enhance the resolution of both P-wave and PS-wave.It can meet the requirement of resolute very thin layers,thus can be used as an effective method for high resolution processing.

Key words: wavelet estimation, frequency improving, deconvolution, reflectance, seismic exploration

中图分类号: