雷达地物回波模糊逻辑识别法的改进及效果检验

地物回波是影响中国新一代天气雷达资料质量的一个非常重要的因素。用模糊逻辑的算法,从识别降水回波和地物回波的特征参数中,选择了效果较好的6个特征参数。根据降水回波与地物回波的特征差异,进行模糊化处理,得出每个象素是地物的可能性,对超出地物阈值的象素点则识别剔除。并针对以前方法对镶嵌在降水中的地物以及小尺度对流云地物过度抑制的问题进行了改进:一是通过在剔除地物回波的基础上增加了回波填补的功能,将降水区域中的"回波空洞"进行有效填补;二是将地物判别阈值由固定的常数变为随距离变化的函数,减小对流云特别是小尺度强对流云边缘地物的过渡抑制。最后用统计平均法检验了整个质量控制算法,通过比较天津雷达质量控制前后的回波强度累加平均图,表明该算法对地物回波有显著的识别效果。

Improvement of the fuzzy logic technique for identifying ground clutter and its verification

Ground clutter is an important factor affecting the radar data quality of the CINRAD. Based on the fuzzy logic algorithm, six efficient parameters from a variety of parameters used in radar quality control algorithms were selected for this study. To remove the bin whose value exceeds a certain threshold, it uses certain fuzzy approach to output a value that quantifies the possibility that each bin will be affected by clutter. In cases where AP echoes are embedded within real rainfall echoes, gaps in rainfall echo areas potentially created by the algorithm will be filled using the echo information above by the improved algorithm. Another aim of this study is to reduce the possibility of misidentification of small-scale strong convective echoes far away from the radar sites by changing each bin’s threshold in the algorithm for different radar distance ranges. The significance of identifying ground clutter has been tested by statistical methods through comparing between the accumulative mean echo intensities before and after the quality control as done by the radar at Tianjin.