基于逐级迭代插值的重力数据扩边方法研究

Research on the Method of Gravity Data Edge Expanding Based on Step By Step Iterative Interpolation

为了研究考虑异常变化特征的扩边插值方法,笔者使用理论重力异常模拟提取的1∶5万重力数据进行研究。通过Surfer软件中7种插值方法的比较,认为径向基函数法在数据扩边时能够获得较好的结果。在此基础上,笔者提出逐级迭代插值的思路并优选扩边参数如下:①核函数为多重二次曲面核函数。②搜索扇区为4个。③搜索点数为64个。④搜索半径R1/R^(2)为6/18,且R^(2)平行于异常整体走向。⑤搜索角度选为当搜索半径长轴R_(2)平行于异常走向时的角度。⑥R^(2)参数一般为0~1。其中,首次扩边时R^(2)参数根据边界点残差对比结果选定,其余各级扩边时R^(2)参数选0.1即可。银额盆地西部LY区块实际资料扩边应用时,采用“三步法”进行插值能够将外扩与内插参数联合考虑,扩边结果显示扩边区与实测区衔接处的重力异常连续光滑,扩边区较好的反映了异常整体趋势和局部变化特征。

In order to study the edge expanding interpolation method considering the variation characteristics of gravity anomaly,we used 1∶50000 gravity data obtained from theoretical gravity anomaly simulation for research.Through the comparison of 7 interpolation methods in surfer software,it is considered that the radial basis function method can obtain better results in data expanding.On this basis,the idea of step by step iterative interpolation is proposed and the expanding parameters are optimized as follows:①The kernel function is multiquadric function,②The number of sectors to research is 4,③The maximum number of data to use from all sectors is 64,④The search radius R 1/R^(2)=6/18,and the radius R2 parallels to the main strike of the anomaly,⑤The search angle was selected when the search radius R^(2)is parallels to the main strike of the anomaly,⑥The R^(2)parameter is generally between 0 and 1,among them,R2 parameter is selected according to the residual error comparison results of boundary points when expanding for the first time,and 0.1 parameter can be selected for other levels of expanding.In the application of actual data edge expanding of LY block in west of Yin’e basin,the“three-step method”can be used for interpolation to jointly consider the external expanding and internal interpolation parameters.The expanding results show that the gravity anomaly at the junction of the expanding area and the measured area is continuous and smooth,and the expanding area better reflects the main strike and local variation characteristics of the anomaly.