一种基于半监督学习的多维条件下电能表误差插值方法  

An error interpolation method for electricity meters under multidimensional conditions based on semi-supervised learning

作　　者：陆煜锌 方彦军[1] Lu Yuxin;Fang Yanjun(School of Electrical Engineering and Automation,Wuhan University,Wuhan,430072,China)

机构地区：[1]武汉大学电气与自动化学院,武汉430072

出　　处：《电测与仪表》2021年第2期153-157,共5页Electrical Measurement & Instrumentation

基　　金：国家自然科学基金资助项目(51807143);中国博士后科学基金特别资助项目(2018T110797);中国博士后科学基金面上项目(2017M612499)。

摘　　要：目前电能表现场检定的研究中有一个关键问题:如何利用相互独立的电能表单维实验误差获得多维实验误差。文中搭建了分别包括温度、谐波、磁场在内的三个独立实验平台,且每个平台均可控制电能表负载参数。设计了虚拟的多维交叉实验点,并提出一种半监督学习方法,实现单维实测误差数据向多维交叉实验数据的插值,获得电能表在多维条件实验下的误差数据。在综合实验平台中验证该方法的有效性和泛化性。At present,there is a key problem in the field verification of electricity meters,which is how to obtain multi-dimensional experimental errors by using the electricity meter error from many unidimensional experiments which are independent of each other.In this paper,three independent experimental platforms,including temperature,harmonics and magnetic field,are built,and each platform can control the load parameters of the electricity meter.Then,many virtual multi-dimensional experimental intersections are designed,and a semi-supervised learning method is proposed to obtain the error data of electricity meter under the multi-dimensional condition experiment.Finally,the effectiveness and generalization of the proposed method are verified in a comprehensive experimental platform.

关 键 词：电能表误差 现场检定 半监督学习 

分 类 号：TM933[电气工程—电力电子与电力传动]