摘要
针对电能质量扰动识别时特征提取不充分和人工提取特征值困难较大,造成最终识别精度不高的问题,本文基于对样本数据进行深度学习提出一种采用卷积神经网络(convolutional neural network,CNN)的电能质量扰动分类算法.文中利用CNN从原始扰动图像中自适应地提取特征并加以分类,同时就三种优化器Adam、Rmsprop、SGD在卷积神经网络中对电能质量扰动进行分类,仿真结果表明Adam、Rmsprop在卷积神经网络中对电能质量扰动识别分类效果要优于SGD.Adam和Rmsprop这两种优化器在电能质量扰动分类中识别速度快,精度高,鲁棒性强.
At present,the feature extraction is not enough and it is difficult to extract the feature value manually,which results in the problem of low recognition accuracy.Based on in-depth learning of sample data,a power quality disturbance classification algorithm using convolutional neural network(CNN)is proposed.CNN is used to extract features from the original disturbed image and classify them.At the same time,three kinds of optimizers,namely,Adam,Rmsprop and SGD are classified in the convolutional neural network.The simulation results show that Adam and Rmsprop are better than SGD in power quality disturbance recognition and classification in convolutional neural network.Adam and Rmsprop optimizers are fast,accurate and robust in power quality disturbance classification.
作者
吴定安
钟建伟
秦勉
向家国
曾凡伟
陈晨
胡凯
WU Dingan;ZHONG Jianwei;QIN Mian;XIANG Jiaguo;ZENG Fanwei;CHEN Chen;HU Kai(School of Information Engineering,Hubei Minzu University,Enshi 445000,China;Enshi Power Supply Company,State Grid Hubei Electric Power Company,Enshi 445000,China)
出处
《湖北民族大学学报（自然科学版）》
CAS
2020年第3期318-321,327,共5页
Journal of Hubei Minzu University:Natural Science Edition
基金
国家自然科学基金项目(61963014)
恩施州科技计划项目(D20180017).
关键词
卷积神经网络
优化器
电能质量扰动
特征值
convolution neural network
optimizer
power quality disturbance
eigenvalue
