本文用一种模式识别方法——拟合优度法研究了大华北地区地震活动的分期。结果表明,拟合优度法与Fisher方法一样,能定量地划分地震活动期;但用后者需采用树状分类器,前者则能一次给出最后的分类结果,而且拟合优度法还可以根据迭代次数来确定合理的分类数。此外,文中对研究区未来时间段是否属于地震高潮期,也给出了判定的方法与结果。 In this paper, we use a good method of pattern recognition-goodness-of-fit method to study the staging of seismicity in the North China. The results show that the goodness-of-fit method, like the Fisher method, can quantitatively divide the seismic activity period. However, the latter should use the tree classifier, the former can give the final classification result at a time, and the goodness-of-fit method According to the number of iterations to determine a reasonable number of categories. In addition, the article also gives the method and result of the judgment whether the future time segment of the study area belongs to the earthquake high tide period.