基于粗糙极化稀疏矩阵的地震属性融合约简方法及其应用  

An approach to seismic attributes fusion and reduction based on rough polarization sparse matrix

作　　者：刘涛平[1] 刘宏杰 娄兵 高新峰 

机构地区：[1]西安交通大学机械工程学院,陕西西安710049 [2]中国石油新疆分公司勘探开发研究院地球物理研究所,新疆乌鲁木齐830013

出　　处：《石油地球物理勘探》2016年第4期774-781,6,共8页Oil Geophysical Prospecting

基　　金：国家自然科学基金青年科学基金项目(41504093);河北省引进留学人员基金课题(C2012003038)联合资助

摘　　要：石油地震勘探中主要采用地震属性进行油气预测,但需针对不同预测对象从众多地震属性中优选敏感属性、剔除冗余属性,以确保预测结果的准确性。为此,在充分研讨粗糙集理论和极化矩阵的基础上,提出一种基于粗糙极化稀疏矩阵的地震属性融合约简方法。通过稀疏极化矩阵组合计算,找到所有的地震属性融合约简;同时还从理论上证明了粗糙极化稀疏矩阵地震属性融合约简方法的可行性。仿真试验和实际资料应用的结果表明:该方法简单、便捷;对地震属性约简处理后,不仅可降低预测的多解性,提高预测精度,而且还可提高预测处理的效率和综合应用效果。In the oil seismic exploration, seismic attributes are mainly adopted for oil and gas prediction. However, too many seismic attributes involved may not lead to a better prediction, and they may lower prediction accuracy instead. Therefore the selection of seismic attributes shall be optimized in accordance with the prediction object. As a result, research on seismic attributes reduction approaches has become a significant part. With the basis of the rough set theory and the polarization matrix, we propose an approach to seismic multi-attribute fusion in this paper. All the seismic attribute reductions are obtained by sparse polarization matrix combination. The feasibility of the proposed approach is further testified in the theory. It is a simple and easy way to access. Simulation tests and practical applications manifest that the proposed approach can not only decrease multiplicity and improve oil and gas prediction accuracy, but also improve prediction efficiency in practice. © 2016, Editorial Department OIL GEOPHYSICAL PROSPECTING. All right reserved.

关 键 词：粗糙集 极化矩阵 地震属性融合 属性约简 油气预测 

分 类 号：P631[天文地球—地质矿产勘探]