摘要
采用多卜勒天气雷达资料,在IBM-PC系列微机上建立强对流天气回波0—2小时临近预报模式。该模式先对PPI图象数据进行预处理,选出一群有预报意义的强回波块,提取它们的一组特征值,然后用模糊聚类法进行分类,并用路径外推法作临近预报,获得比较客观的预报结果。
A 0—2 hour nowcasting model of the strong-convection echo system is established on the IBM-PC microcomputer system in terms of the weather Doppler radar dataset. From the PPI images preprocessed with the model proposed, a group of patches of intense echoes of interset is singled out, from Which a set of eigenvalues is extracted and classified using the fuzzy cluster technique. With the aid of extrapolation of the track a nowcasting is prepared, whereby good results have been obtained on an objective basis.
出处
《南京气象学院学报》
CSCD
1993年第3期302-307,共6页
Journal of Nanjing Institute of Meteorology
