基于小波Adaline改进的充电桩谐波分析及扰动定位方法  

Harmonic wave analysis and disturbance localization method of charging pile based on wavelet Adaline improvement

作　　者：王晨 彭家强 王淑侠[2] 邬蓉蓉 王世刚 吴其琦[1] WANG Chen;PENG Jiaqiang;WANG Shuxia;WU Rongrong;WANG Shigang;WU Qiqi(School of Automation,Guangxi University of Science and Technology,Liuzhou 545006,China;Mechanical and Electrical College,Northwestern Polytechnical University,Xi’an 710072,China;Electric Power Research Institute,Guangxi Power Grid Co Ltd,Nanning 530023,China)

机构地区：[1]广西科技大学自动化学院,广西柳州545006 [2]西北工业大学机电学院,陕西西安710072 [3]广西电网有限责任公司电力科学研究院,广西南宁530023

出　　处：《传感器与微系统》2024年第6期129-132,共4页Transducer and Microsystem Technologies

基　　金：国家重点研发计划资助项目(2019YFB1703800);广西自然科学基金资助项目(2018GXNSFAA050-029);广西自然科学基金资助项目(2018GXNSFAA294085)。

摘　　要：针对Adaline网络在复杂信号谐波检测中基波频率难以获取的问题,提出一种小波重构与改进型Adaline网络结合的谐波分析方法,并将该方法应用于充电桩谐波分析中。首先,用Mallat算法对原信号分解,再将细节信号重构实现扰动定位,将近似信号重构得到去噪原信号与基波信号,通过基波获取频率。其次,对Adaline网络相位提取算法进行优化,使其兼容全部相位范围,将去噪原信号通过优化后的Adaline网络检测谐波幅值及其对应的相位,综合误差为0.766%,比传统的Adaline模型精度提升了28.642%;最后,将该方法应用于充电桩模型,实现谐波检测,并精准定位扰动发生时段,进一步论证了该方法的适用性。Aiming at the problem that it is difficult to obtain the fundamental wave frequency of Adaline network in harmonic wave detection of complex signals,a harmonic wave analysis method combining wavelet reconstruction and improved Adaline network is proposed and applied to harmonic analysis of charging piles.Firstly,Mallat algorithm is used to decompose the original signal,and then the detail signal is reconstructed to achieve disturbance localization.The approximate signal is reconstructed to obtain the denoised original signal and fundamental wave signal,and the frequency is obtained through the fundamental wave.Secondly,the phase extraction algorithm of Adline network is optimized to make it compatible with all phase ranges.The denoised original signal is detected for harmonic wave amplitude and its corresponding phase through the optimized Adaline network.The comprehensive error is 0.766%,which is 28.642%higher than the precision of the traditional Adaline model.Finally,the method is applied to the charging pile model to realize harmonic wave detection and accurately locate the disturbance occurrence period,further demonste the applicability of the method.

关 键 词：谐波检测 MALLAT算法 ADALINE网络 充电桩 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TM935[自动化与计算机技术—计算机科学与技术]