基于关联规则的多普勒天气雷达故障挖掘算法分析

基于辽宁丹东多普勒天气雷达等台站,针对当前多普勒天气雷达发射机、接收机、伺服等软硬件故障挖掘过程中,受冗余、缺失、错误数据的影响导致最终挖掘率和挖掘准确率不高,误挖率以及漏挖率始终较高的问题,提出了基于关联规则的多普勒天气雷达故障挖掘算法.该算法利用关联规则挖掘技术原理以及相关定义定理,将收集到的多普勒天气雷达故障数据集划分成不同区域,滤除弱关联故障数据,保留强关联故障数据,完成对多普勒天气雷达故障数据的预处理,并构建了故障原始判断表和故障挖掘模型,在此基础上,利用故障原始判断表和挖掘模型对多普勒天气雷达故障进行定位和类别判断,完成挖掘.性能测试结果显示,采用所提算法能够实现多种类型多普勒天气雷达故障的高精度挖掘,为多普勒天气雷达系统的安全、稳定运行提供了有效保障.

Fault Mining Algorithm Analysis of Doppler Weather Radar Based on Association Rules

Based on Liaoning Dandong Doppler weather radar and other stations , for the current Doppler weather radar transmitter , receiver , servo and other software and hardware faults , the mining process is affected by redundant , missing , and erroneous data , resulting in the low final mining rate and mining accuracy．And the rate of mis-mining and missing-mining is always high．A?Doppler weather radar? ?fault mining algorithm basedon association rules is proposed．This algorithm uses the principles of association rule mining technology and relateddefinition theorems to collect．The Doppler weather radar fault data set is divided into different regions , filtering outweakly correlated fault data , retaining strong correlated fault data , completing preprocessing of Doppler weatherradar fault data , and constructing the original fault judgment table and fault mining model.Based on this , the original fault judgment table and mining model are used to locate and classify Doppler weather radar faults and complete mining．Performance test results show that the proposed algorithm can achieve high-precision mining of various types of Dopplerweather radar faults , provides effective guarantee for the safe and stable operation of the Doppler weather radar system．