软距离计算的地震聚类方法*

传统的基于真实距离的聚类分析方法不利于地震不同断层破裂传播和愈合速度的精确计算。为提高地震预测精度，提出并建立了基于软距离计算的聚类方法。给出了基于软距离聚类过程、软距离计算方法以及具体的基于软距离计算的聚类算法。以现实的强震样本点作为聚类数据源，采用该聚类方法以及其它传统聚类方法对该样本数据进行聚类分析。分析结果表明，采用该聚类方法获得的聚类中心点更接近地壳应力场演变的客观真实性，该聚类分析方法为地震的断层带下次发生强震的精确计算提供了很好的计算依据。

Earthquake clustering method based on soft distance calculation

Traditional cluster analysis based on the true distance method is not conducive to the earthquake fault rupture propagation and different healing accurate calculation. This paper proposed and gave the new clustering method based on soft distance calculation. The clustering process based on soft distance calculation, calculation method for soft distance and the specific clustering algorithm based on soft distance are given. For the reality of the earthquake cluster as a data source, to use this clustering method and other traditional clustering method to cluster analysis for the same data source, analysis results show that the clustering method obtained cluster center with the earth stress field evolution have the same center, which this method has objective truth. The cluster analysis method for the earthquake fault zone in the accurate calculation of the next strong earthquake provides a good basis for the calculation.