基于协克里金的雨量雷达融合置信处理方法  

A fusion-based confidence processing algorithm for regional rainfall radar based on Co-Kriging

作　　者：陈根华[1] 莫正威 CHEN Genhua;MO Zhengwei(School of Information Engineering,Nanchang Institute of Technology,Nanchang 330099,China)

机构地区：[1]南昌工程学院信息工程学院,江西南昌330099

出　　处：《南昌工程学院学报》2022年第3期64-71,共8页Journal of Nanchang Institute of Technology

基　　金：江西省教育厅科学技术研究项目(GJJ170990);国家自然科学基金资助项目(61401187);江西省重点研发计划项目(20212BBG73009)。

摘　　要：针对雨量雷达的“测不准”和雨量计的“测不到”等问题,提出了一种雨量雷达反射率因子的融合置信处理算法,并定量分析了雨量雷达降雨数据的相对置信度。首先对雨量雷达反射率因子数据进行Z-score标准化,再以标准化后的雷达反射率因子数据为辅变量,以雨量计雨量数据为主变量,通过协克里金方法实现数据的融合处理,实现雨量计数据与雨量雷达数据的融合。该算法在数据层实现了雨量计雨量数据与雨量雷达反射率因子数据的融合,并利用相对置信度分析了算法的适用性与稳健性。实测数据与仿真试验验证了融合处理算法的有效性与正确性,为工程应用上雨量雷达探测降雨提供了理论指导。In view of the“imprecision”of rainfall radar and the“detecting unavailability”of rain gauge,a fusion-based confidence processing algorithm for regional rainfall radar is proposed.The concept of the relative confidence degree(RCD)is proposed for quantitative analysis of calibration of rainfall radar data.The radar reflectivity factor of rainfall radar is preprocessed by Z-score normalization.Then the fusion-based confidence processing for radar reflectivity factor of rainfall radar is achieved by Co-Kriging algorithm with the preprocessed data as the auxiliary variable and conventional rain gauge data as the main variable,which implements quantitative precipitation estimation with the radar reflectivity factor data and rain gauge data in data layer.The data fusion between rain gauge and rainfall radar is accomplished for reducing the variability of rainfall radar data.Real data and simulation results demonstrate the effectiveness and correctness of the fusion-based confidence processing algorithm.

关 键 词：区域雨量雷达 融合 雨量计 协克里金法 相对置信度 

分 类 号：TN95[电子电信—信号与信息处理] P426.6[电子电信—信息与通信工程]