摘要
针对传统电压暂降源定位方法准确率低的缺点,提出了基于希尔伯特-黄变换(Hilbert-Huang transform,HHT)HHT和GA-BP的定位方法。用集合经验模态分解(ensemble empirical mode decomposition,EEMD)对故障期间的电压和电流进行处理,得到有效特征值-电流实部Icosθ和系统轨迹斜率k,接着用GA-BP神经网络对有效特征值进行分类找出故障源的初步位置,然后用飞蛾扑火优化(moth-flame optimization,MFO)算法对故障线路的电压电流构建的数学模型求解,从而对电压暂降源进行精确定位。同时,根据三相电压幅值之间的关系区分故障类型。最后,以双电源系统仿真模型为例对所提方法进行验证,仿真结果表明:提出的定位方法准确率和精度很高,能够准确地定位出电压暂降源的位置。
Aiming at the low accuracy of traditional voltage sag source positioning methods,this paper proposes a new location method based on HHT and GA-BP.Firstly,the EEMD(ensemble empirical mode decomposition)Hilbert-Huang Transform(HHT)is used to process the voltage and current during the fault period,and the effective eigenvalues-the current real part Icosθand system trajectory slope k are obtained.Then the GA-BP neural network is used to classify the effective eigenvalues to get the initial location of the voltage sag source.And then,the moth to flame optimization(MFO)algorithm is used to solve the mathematical model of voltage and current of the fault line,so that the precise location of the voltage sag source is obtained.At the same time,the fault types are distinguished according to the relationship between the three-phase voltage amplitudes.Finally,a simulation model of a dual power supply system is used to verify the proposed method,and the simulation results show that the proposed method has high accuracy and precision in positioning and can precisely locate the voltage sag source.
作者
杨桢
马钰超
李丽
李鑫
马子莹
YANG Zhen;MA Yuchao;LI Li;LI Xin;MA Ziying(Faculty of Electrical and Control Engineering,Liaoning Technical University,Xingcheng 125105,China;Fuxin Power Supply Company of State Grid Liaoning Power Co.,Ltd.,Fuxin 123000,China;Tangshan Fengnan District Power supply Branch of State Grid Jibei Electric Power Co.,Ltd.,Tangshan 063000,China)
出处
《中国电力》
CSCD
北大核心
2022年第3期97-104,共8页
Electric Power
基金
国家自然科学基金资助项目(深部复合煤岩卸荷破裂热红外辐射机理及多场耦合模型研究,51604141)。
关键词
集合经验模态分解
希尔伯特变换
GA-BP神经网络
电流实部
系统斜率轨迹
ensemble empirical mode decomposition
Hilbert transformation
GA-BP neural network
real part of current
the slope trajectory of the system
