自适应正则化滤波化极方法研究  

Research on the adaptive regularization filtering method of reduction-to-the-pole

作　　者：何涛 王皓 张义蜜 王万银[1,2,3] Colin G.Farquharson J.Kim Welford HE Tao;WANG Hao;ZHANG YiMi;WANG WanYin;Farquharson Colin G;Kim Welford J(College of Geology Engineering and Geomatics,Chang'an University,Xi'an 710054,China;National Engineering Research Center of offshore Oil and Gas Exploration,Beijing 100028,China;Key Laboratory of Marine Geology and Environment,Chinese Academy of Sciences,Qingdao Shandong 266071,China;Department of Earth Sciences,Memorial University of Newfoundland,St.John's,NewfoundlandAlB 3X5,Canada)

机构地区：[1]长安大学地质工程与测绘学院,西安710054 [2]海洋油气勘探国家工程研究中心,北京100028 [3]中国科学院海洋地质与环境重点实验室,山东青岛266071 [4]Department of Earth Sciences,Memorial University of Newfoundland,St.John's,AlB 3X5,Canada

出　　处：《地球物理学报》2024年第3期1255-1272,共18页Chinese Journal of Geophysics

基　　金：国家重点研发计划课题(2017YFC0602202);中海石油(中国)有限公司科技项目课题“中国近海潜在富油凹陷深部构造差异性”;海洋油气勘探国家工程研究中心2022年开放基金课题“中国近海圈层构造和区域构造对含油气盆地和凹陷控制机理研究”联合资助。

摘　　要：化极是消除磁力异常斜磁化影响的关键步骤,可以在波数域(频率域)和空间域中实现.相比于空间域,在波数域(频率域)进行化极简单且高效,但在低纬度地区存在不稳定性问题,因此明确低纬度地区的范围,研究稳定、实用、精度高的低纬度地区化极方法具有重要意义.本文通过理论研究得出化极因子振幅的最大值为2时,对应的磁化倾角的值为±45°,当磁化倾角在该范围内时,引起化极不稳定.因此采用正则化思想,提出了自适应正则化滤波化极方法(ARF-RTP),该方法利用化极因子振幅构建正则化滤波函数,并以振幅值等于2作为约束计算化极方法的参数值,从而实现自适应化极处理,提高了化极方法的稳定性及实用性.同时结合迭代算法提高了此方法的精度,有效地解决了低纬度地区化极的不稳定性问题.自适应正则化滤波这一思想不但可以解决化极的不稳定性问题,而且可以解决分量转换或者磁化方向转换的不稳定性问题以及类似数据处理中的不稳定性问题,具有广阔的推广应用前景.Reduction-to-the-pole (RTP) is a significant step in eliminating the effects of oblique magnetization on magnetic anomaly. This process can be conducted in both the wave-number (frequency domain) and the spatial domains. The RTP in the wave-number domain (frequency domain) is more efficient than in the spatial domain, but it can be unstable in low latitude areas. Thus, it is necessary to define the scope of the low latitude areas and develop stable, practical, and high-precision RTP methods. In this paper, the theoretical study shows that the maximum value of the amplitude is 2, and the corresponding magnetization inclination is ±45°. Within this range of magnetization inclination, causing the instability in RTP procedure. To address this issue, an adaptive regularization filtering method of reduction-to-the-pole (ARF-RTP) is proposed based on the regularization idea. The method constructs a regularization filtering function using the amplitude of RTP factor, and with the amplitude value of 2 as the constraint, the determination value of the parameter in the method is calculated to realize the adaptive RTP of magnetic anomaly, which improves the stability and practicability of the method. Additionally, an iterative algorithm is used to improve the accuracy of the method. The proposed ARF-RTP method effectively solves the instability problem of the RTP in the low latitude areas. The accuracy, stability, and practicality of the method are verified using synthetic models and real data. The adaptive regularization filtering can not only solve the instability problem of RTP but also address instability problems in data processing, such as component transformation or magnetization direction transformation of magnetic field. The ARF-RTP method has broad application prospects.

关 键 词：化极 低纬度地区 振幅 自适应正则化滤波 迭代算法 

分 类 号：P631[天文地球—地质矿产勘探]