一种基于经验模态分解(EMD)的位场分离方法  

A potential field separation method based on Empirical Mode Decomposition

作　　者：曾琴琴[1,2] 刘天佑[1,2] 

机构地区：[1]中国地质大学地球物理与空间信息学院,湖北武汉430074 [2]中国地质大学构造与油气资源教育部重点实验室,湖北武汉430074

出　　处：《石油地球物理勘探》2010年第6期914-917,共4页Oil Geophysical Prospecting

基　　金：中石化海相前瞻性研究项目"中国南方(中上扬子)大陆构造与海相油气前景"(编号YPH08002)资助

摘　　要：经验模态分解(EMD)是一种利用信号极值特征进行时空滤波的方法。经分解得到不同阶次固有模态函数(包含从高到低不同频率段的成分),而残余量为信号的趋势分量,代表信号的平均趋势。本文将经验模态分解用于地球物理场的位场分解中,提出了一种基于EMD的位场分离方法。与常规位场分离方法中的滑动平均法和趋势分析法相比,EMD无须预先设置参数,因而不受人为因素的影响,其结果能够更好地反映位场的固有特征。应用该法对新疆准噶尔盆地准东地区的磁异常资料进行处理,从中识别出与石炭系油气有关的火成岩异常。The Empirical Mode Decomposition(EMD)proposed by Norden E Huang is a kind of spatial and temporal filtering method which takes the signal extremum characteristics scales into account.The intrinsic mode functions of different scales contain various bands of compositions with frequencies from high to low,and the residual value is the trend component that representing the average trend of the signal.Therefore,a kind of potential field separation method that based on EMD was proposed in this article.Compared with the conventional methods,such as the moving average method and the trend analysis method in the potential field separation method,there are some advantages for EMD,for example,we don't need to set any parameters beforehand,so there are less man-made influences,moreover,the results could better reflect the potential field intrinsic characteristics.The method was used in the magnetic anomaly data processing in the east area of the Junggar Basin,the igneous rock anomaly which is related to the Carboniferous oil&gas was identified from the processed data.

关 键 词：经验模态分解(EMD) 固有模态函数(I MF) 位场分离 油气勘探 准噶尔盆地 火成岩油气藏 

分 类 号：P631[天文地球—地质矿产勘探]