几种异常值判别准则在气象计量数据处理中的应用  

Application of some statistical criteria in meteorological data processing

作　　者：李常春[1] 常克昊 刘禹杉 Li Changchun;Chang Kehao;Liu Yushan(Jiangxi Provincial Atmospheric Observation and Technical Center,Nanchang 330096,China;Meteorological Station of Flight Test Station of Changhe Aircraft Industry(Group)Co.,Ltd.,Jingdezhen 330000,China;Jiangxi Provincial Meteorological Service Center,Nanchang 330096,China)

机构地区：[1]江西省大气探测技术中心,南昌330096 [2]航空工业集团昌河飞机工业(集团)有限责任公司试飞站气象台,景德镇330000 [3]江西省气象服务中心,南昌330096

出　　处：《电子测量技术》2020年第23期68-72,共5页Electronic Measurement Technology

基　　金：国家自然科学基金(42065011);江西省气象科技重点项目(赣气发[2018]127号)资助。

摘　　要：针对在判定在气象计量数据异常值时缺乏相关研究的现状,首先,详细的介绍了3种常规的异常值判定准则,给出了各自的适用性条件;然后,通过实例,并结合各准则的适用性条件进行了实际应用;结果表明,3种判定准则可以有效的对气象计量领域的测量数据进行异常值的判定;在进行实际应用时,可根据测量数据的多少、计算的复杂性以及是否需要判定多个异常值等实际情况来选择具体的判定准则。异常值判别准则在气象计量数据中的合理应用有利于提升气象探测数据的准确性和可靠性。In view of the lack of relevant research in judging the abnormal value of meteorological measurement data, Firstly, introduces three kinds of conventional discriminant criteria of abnormal value in detail, and gives their respective applicability conditions;Then, the practical application is carried out by examples and the applicability of each criterion;The results show that the three discriminant criteria can effectively judge the abnormal values of the measurement data in the field of meteorological measurement. In practical applications, specific discriminant criteria can be selected according to the actual conditions such as the amount of measurement data, the complexity of calculation, and whether multiple abnormal values need to be judged. The reasonable application of the discriminant criterion of abnormal value in meteorological measurement data is helpful to improve the accuracy and reliability of meteorological observation data.

关 键 词：计量 气象 异常值 判别准则 

分 类 号：TP2[自动化与计算机技术—检测技术与自动化装置]