基于改进的小波神经网络充电站谐波检测  

Charging station harmonic detection based on improved wavelet neural network

作　　者：杨乐 帕孜来·马合木提[1] YANG Le;MAHEMUTI Pazilai(School of Electrical Engineering,Xinjiang University,Urumchi 830049,China)

机构地区：[1]新疆大学电气工程学院,新疆乌鲁木齐830049

出　　处：《现代电子技术》2021年第5期156-160,共5页Modern Electronics Technique

基　　金：国家自然科学基金资助项目(61364010);新疆维吾尔自治区自然科学基金(2016D01C038)。

摘　　要：大量的充电桩接入电网引起了谐波问题,严重影响了电网的电能质量,为了给谐波补偿提供可靠的依据,能够快速准确地检测出注入电网的谐波含量是非常必要的。首先对充电桩注入电网的谐波特点进行分析,然后提出基于改进小波神经网络算法,能够快速并准确地检测出充电桩引起的谐波,最后使用Matlab的仿真平台进行仿真验证。仿真结果表明,基于改进的小波神经网络可以精确检测出充电桩向电网注入谐波的幅值和相位,也可以有效地降低迭代次数,满足电动汽车充电站谐波检测要求。A number of charging piles are connected to the power grid,which causes harmonic problems and seriously affect the power quality of the power grid.In view of this,it is very necessary to quickly and accurately detect the harmonic content injected into the power grid,so as to provide a reliable basis for harmonic compensation.The characteristics of harmonics injected into the power grid by the charging piles are analyzed.And then,an improved wavelet neural network algorithm is proposed,which can detect the harmonics caused by the charging pile quickly and accurately.Finally,the simulation platform Matlab is used for verification.The simulation results show that the improved wavelet neural network can not only accurately detect the amplitude and phase of the harmonics injected into the power grid by the charging pile,but also reduce the iterations effectively,which meet the requirements of harmonic detection for charging stations of electric vehicles.

关 键 词：充电桩 非线性负载 电网 仿真 谐波 Matlab 小波变换 神经网络 电能质量 

分 类 号：TN06-34[电子电信—物理电子学] TM711[电气工程—电力系统及自动化]