复杂电磁环境下基于球谐函数分解的磁场传感数据反演电流方法  

Novel Current Inversion Method of Magnetic Field Sensing Data Based on Spherical Harmonics Decomposition in Complex Electromagnetic Environment

作　　者：李鹏[1,2] 马浩宇 田兵 刘仲 吕前程 胡军 LI Peng;MA Haoyu;TIAN Bing;LIU Zhong;LÜQiancheng;HU Jun(Tsinghua University(Department of Electrical Engineering)-China Southern Power Grid Digital Grid Research Institute Co.,Ltd.Joint Research Center for Smart Sensor and Transparent Grid,Beijing 100084,China;China Southern Power Grid Digital Grid Research Institute,Guangzhou 510663,China;Department of Electrical Engineering,Tsinghua University&State Key Lab of Operation and Control of New-generation Power System,Beijing 100084,China)

机构地区：[1]清华大学(电机系)–南方电网数字电网研究院智能传感与透明电网联合研究中心,北京100084 [2]南方电网数字电网研究院,广州510663 [3]清华大学电机系及新型电力系统运行与控制全国重点实验室,北京100084

出　　处：《高电压技术》2023年第6期2365-2373,共9页High Voltage Engineering

基　　金：国家自然科学基金(52125703);南方电网数字电网研究院科技项目(2100002020030103SJ01718);四川省科技计划项目(2021YFG0348)。

摘　　要：通过对空间磁场信息进行反演,能够获取复杂电磁环境下的被测电流信息。针对电气装备中的电流测量需求,提出一种基于球谐函数分解的电流反演方法。该方法基于空间磁场数据,采用球谐函数对干扰磁场源进行解析分解,从而在干扰磁场源参数未知的情况下,实现被测线电流源及干扰磁场源的解耦。分别通过弹性网正则化及贝叶斯弹性网正则化方法进行电流反演,在干扰源的磁场强度占被测线电流产生磁场强度的60%时,得到最大不超过10%的线电流反演误差,并分析了球谐函数分解的最大分解次数选取原则。所提出的非介入式电流测量新方法克服了复杂电磁环境下的外部干扰问题,适用于电力物联网广泛的电流测量计算需求。By inverting the spatial magnetic field information,it is possible to acquire the current parameters in complex electromagnetic environment.Consequently,we proposed a novel current inversion method based on spherical harmonics decomposition.In this method,spatial magnetic field data are utilized and spherical harmonic functions are applied to analyze and decompose the interfering magnetic field sources,thus achieving decoupling between the measured line current source and the interfering magnetic field source when the parameters of the interfering magnetic field source are unknown.The current inversion is performed using both elastic net regularization and Bayesian elastic net regularization methods.The results indicate that,in the situation where the ratio of interference field to the field of interest reaches 60%,the error of the calculated current is significantly decreased to no more than 10%.Moreover,the selection principle for the maximum decomposition order of spherical harmonic function decomposition is also analyzed.The proposed non-intrusive method for current measurement overcomes external interference issues in complex electromagnetic environments and is suitable for widespread current measurement and calculation requirements in the electric power Internet of Things(IoT).

关 键 词：电流测量 球谐函数分解 弹性网正则化 干扰解耦 电流反演 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] TM933.1[自动化与计算机技术—控制科学与工程]