摘要
在电网运行过程中,强烈的噪声干扰会显著降低电能质量扰动的识别精度。为有效解决这一问题,提升电能质量扰动识别的准确性,本文提出一种电能质量扰动识别方法,该方法基于深度残差网络,可实现单一扰动与复合扰动的识别。该方法可通过数据增强、优化网络结构、运用正则化与批归一化、改进激活函数、多尺度特征融合来增强抗噪性。仿真实验结果表明:本文提出的方法能够在严重噪声干扰条件下,迅速且准确地对电能质量复合扰动进行辨识。
The interference of strong noise in the power grid can seriously affect the identification of power quality disturbances.To improve the accuracy of power quality disturbance identification,this paper proposes a composite power quality disturbance identification method based on the deep residual network.This method can enhance the noise resistance through data augmentation,network structure optimization,regularization and batch normalization techniques,improvement of activation functions,and multi-scale feature fusion.The simulation experiment results show that the proposed method can quickly and accurately identify composite power quality disturbances under the interference of strong noise.
作者
王岩
张红旗
石福利
Wang Yan;Zhang Hongqi;Shi Fui(Inner Mongolia Agricultural University,Hohhot,China)
出处
《科学技术创新》
2025年第20期75-79,共5页
Scientific and Technological Innovation
关键词
电能质量
扰动识别
深度学习
残差网络
power quality
disturbance identification
deep learning
residual network
