基于层次聚类分析的离散欧拉解质量控制方法

Quality Control of Dispersed Euler Solutions Using Hierarchical Clustering Analysis

欧拉反褶积法采用滑动欧拉窗口方式求解场源相关信息,该处理方法的解算结果是一群离散欧拉解,如何对这些离散欧拉解进行质量控制已成为应用欧拉反褶积法的瓶颈问题。针对场源构造指数和深度变化规律,给出了发散解滤波模型,根据欧拉解空间位置分布特点,提出了层次聚类分析方案对离散欧拉解进行质量控制。采用磁性球体和长方体模型仿真试验对所提方法应用效果进行了验证,实验结果表明:滤波措施能消除大部分无效离散欧拉解,结合层次聚类分析方案,实现了场源参数准确确定。

Shifted Euler window is fluently used in Euler deconvolution interpretation to calculate the source information,which results in a series of dispersed Euler solutions.How to implement quality control with those dispersed Euler solutions has become a bottleneck problem in applying Euler deconvolution approach.The filtering model is given on the basis of structural index and depth variety characteristic.Adopting hierarchical clustering method,the dispersed Euler solutions are isolated to different clusters.The effectiveness of the suggested techniques has been illustrated by simulation example.The conclusion shows that numerous invalid Euler solutions could be eliminated with filtering measure,and combining hierarchical clustering analysis means,the source parameters are exactly determinated.