电子式电流互感器畸变信号自动识别方法仿真  

Simulation of Automatic Recognition Method for Distortion Signal of Electronic Current Transformer

作　　者：高学军[1] 童世成 GAO Xue-jun;TONG Shi-cheng(China Three Gorges University,College of Electricity and New Energy,Yichang Hubei 443002,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002

出　　处：《计算机仿真》2020年第8期324-327,336,共5页Computer Simulation

基　　金：湖北省微电网工程技术研究中心开放基金(2015KDW03)。

摘　　要：为了提高畸变信号自动识别的准确性,提出基于小波变换的电子式电流互感器畸变信号自动识别方法。构建小波阈值神经网络模型,将小波最优阈值去噪器融入神经网络模型中,对电子式电流互感器中含噪信号进行去噪,并利用小波变换模极大值理论识别畸变信号的奇异点,获取信号的Lipschitz指数、小波能量系数、均差值、形成畸变信号的特征空间,根据随机生成的畸变信号样本在特征空间的分布,形成畸变信号的诊断模型,实现电子式电流互感器畸变信号自动识别。实验结果表明,所提方法能够快速、准确识别电子式电流互感器中的畸变信号,且具有一定的可行性。In order to improve the accuracy of automatic recognition for distortion signal,a method of automatically recognize the distortion signal of electronic current transformer based on wavelet transform was proposed.Firstly,the neural network model based on wavelet threshold was constructed and the optimal wavelet threshold denoiser was integrated into the neural network model.Secondly,the noisy signal in electronic current transformer was denoised.On this basis,the wavelet transform modulus maximum theory was used to recognize the singularity of distorted signal and thus to get Lipschitz exponent,wavelet energy coefficient,the mean difference and the feature space of distorted signal.According to the distribution of randomly generated distorted signal samples in feature space,the diagnostic model of distorted signal was built.Thus,the automatic recognition of distorted signal in electronic current transformer was achieved.Simulation results show that the proposed method can quickly and accurately recognize the distortion signal in electronic current transformer.This method has certain feasibility.

关 键 词：电子式电流互感器 畸变信号 自动识别方法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]